essay decision support system

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Review Article
• Open access
• Published: 06 February 2020

An overview of clinical decision support systems: benefits, risks, and strategies for success

• Reed T. Sutton   ORCID: orcid.org/0000-0002-3009-1914 1 ,
• David Pincock 2 ,
• Daniel C. Baumgart 1 ,
• Daniel C. Sadowski 1 ,
• Richard N. Fedorak 1 &
• Karen I. Kroeker 1  

npj Digital Medicine volume  3 , Article number:  17 ( 2020 ) Cite this article

371k Accesses

1029 Citations

114 Altmetric

Metrics details

• Drug regulation
• Health services
• Medical imaging

Computerized clinical decision support systems, or CDSS, represent a paradigm shift in healthcare today. CDSS are used to augment clinicians in their complex decision-making processes. Since their first use in the 1980s, CDSS have seen a rapid evolution. They are now commonly administered through electronic medical records and other computerized clinical workflows, which has been facilitated by increasing global adoption of electronic medical records with advanced capabilities. Despite these advances, there remain unknowns regarding the effect CDSS have on the providers who use them, patient outcomes, and costs. There have been numerous published examples in the past decade(s) of CDSS success stories, but notable setbacks have also shown us that CDSS are not without risks. In this paper, we provide a state-of-the-art overview on the use of clinical decision support systems in medicine, including the different types, current use cases with proven efficacy, common pitfalls, and potential harms. We conclude with evidence-based recommendations for minimizing risk in CDSS design, implementation, evaluation, and maintenance.

Similar content being viewed by others

The potential of artificial intelligence to improve patient safety: a scoping review

Why do probabilistic clinical models fail to transport between sites

Effect of genetics clinical decision support tools on health-care providers’ decision making: a mixed-methods systematic review

Introduction: what is a clinical decision support system.

A clinical decision support system (CDSS) is intended to improve healthcare delivery by enhancing medical decisions with targeted clinical knowledge, patient information, and other health information. 1 A traditional CDSS is comprised of software designed to be a direct aid to clinical-decision making, in which the characteristics of an individual patient are matched to a computerized clinical knowledge base and patient-specific assessments or recommendations are then presented to the clinician for a decision. 2 CDSSs today are primarily used at the point-of-care, for the clinician to combine their knowledge with information or suggestions provided by the CDSS. Increasingly however, there are CDSS being developed with the capability to leverage data and observations otherwise unobtainable or uninterpretable by humans.

Computer-based CDSSs can be traced to the 1970s. At the time, they had poor system integration, were time intensive and often limited to academic pursuits. 3 , 4 There were also ethical and legal issues raised around the use of computers in medicine, physician autonomy, and who would be at fault when using the recommendation of a system with imperfect ‘explainability’. 5 Presently, CDSS often make use of web-applications or integration with electronic health records (EHR) and computerized provider order entry (CPOE) systems. They can be administered through desktop, tablet, smartphone, but also other devices such as biometric monitoring and wearable health technology. These devices may or may not produce outputs directly on the device or be linked into EHR databases. 6

CDSSs have been classified and subdivided into various categories and types, including intervention timing, and whether they have active or passive delivery. 7 , 8 CDSS are frequently classified as knowledge-based or non-knowledge based. In knowledge-based systems, rules (IF-THEN statements) are created, with the system retrieving data to evaluate the rule, and producing an action or output 7 ; Rules can be made using literature-based, practice-based, or patient-directed evidence. 2 CDSS that are non-knowledge based still require a data source, but the decision leverages artificial intelligence (AI), machine learning (ML), or statistical pattern recognition, rather than being programmed to follow expert medical knowledge. 7 Non-knowledge based CDSS, although a rapidly growing use case for AI in medicine, are rife with challenges including problems understanding the logic that AI uses to produce recommendations (black boxes), and problems with data availability. 9 They have yet to reach widespread implementation. Both types of CDSS have common components with subtle differences, illustrated in Fig. 1 .

They are composed of (1) base: the rules that are programmed into the system (knowledge-based), the algorithm used to model the decision (non-knowledge based), as well as the data available, (2) inference engine: takes the programmed or AI-determined rules, and data structures, and applies them to the patient’s clinical data to generate an output or action, which is presented to the end user (eg. physician) through the (3) communication mechanism: the website, application, or EHR frontend interface, with which the end user interacts with the system 9 .

CDSS have been endorsed by the US Government’s Health and Medicare acts, financially incentivizing CDS implementation into EHRs. 10 In 2013, an estimated 41% of U.S. hospitals with an EHR, also had a CDSS, and in 2017, 40.2% of US hospitals had advanced CDS capability (HIMSS Stage 6). 11 Elsewhere, adoption rates of EMRs have been promising, with approximately 62% of practitioners in Canada in 2013. 12 Canada has had significant endorsement from the government level, as well as Infoway, a not-for-profit corporation. 13 England has also been a world leader in healthcare IT investment, with up to 20 billion euros invested back in 2010. 13 Several countries have also managed to implement national health records, at least for patient-facing data, including Denmark, Estonia, Australia, and others. 14

The scope of functions provided by CDSS is vast, including diagnostics, alarm systems, disease management, prescription (Rx), drug control, and much more. 15 They can manifest as computerized alerts and reminders, computerized guidelines, order sets, patient data reports, documentation templates, and clinical workflow tools. 16 Each CDSS function will be discussed in detail throughout this review, with the potential and realized benefits of these functions, as well as unintended negative consequences, and strategies to avoid harm from CDSS. Methodology used to inform the review is shown in Box 1 .

Box 1. Methods and sources used for this overview

MEDLINE search 1980-January 2018. Key words: CDSS, diagnostic decision support system/DDSS, personal health record/PHR decision support, EHR decision support

Hand searches of the references of retrieved literature

University libraries searching for texts on clinical decision support systems and other keywords mentioned above

Personal and local experience working with healthcare technology and decision support systems

Functions and advantages of CDSS

Patient safety.

Strategies to reduce medication errors commonly make use of CDSS (Table 1 ). Errors involving drug-drug interactions (DDI) are cited as common and preventable, with up to 65% of inpatients being exposed to one or more potentially harmful combinations. 17 CPOE systems are now designed with drug safety software that has safeguards for dosing, duplication of therapies, and DDI checking. 18 The types of alerts generated by these systems are among the most disseminated kind of decision support. 19 However, studies have found a high level of variability between how alerts for DDIs are displayed (e.g., passive or active/disruptive), which are prioritized, 20 , 21 and in the algorithms used to identify DDIs. 18 , 22 Systems often have varying degrees of irrelevant alerts presented, and there is no standard for how best to implement which alerts to providers. The US Office of the National Coordinator for Health Information Technology has developed a list of ‘high-priority’ list of DDIs for CDS, which has reached various levels of endorsement and deployment in CDSS’ of other countries including the U.K., Belgium, and Korea. 20 , 21 , 23

Other systems targeting patient safety include electronic drug dispensing systems (EDDS), and bar-code point-of-care (BPOC) medication administration systems. 24 These are often implemented together to create a ‘closed loop’, where each step of the process (prescribing, transcribing, dispensing, administering) is computerized and occurs within a connected system. At administration, the medication is automatically identified through radio-frequency identification (RFID) or barcodes and crosschecked with patient information and prescriptions. This presents another target for CDSS and the potential benefit is the prevention of medication administration errors occurring at the ‘bedside’ (opposed to further upstream). Adoption is relatively low, partly due to high technology requirements and costs. 25 However; studies show good efficacy for these systems in reducing errors. 26 Mohoney et al. showed that many of these systems can be combined with CPOE and CDSS simultaneously, with reduced prescribing error rates for drug allergy detection, excessive dosing, and incomplete or unclear ordering. 24 As with most CDSS, errors can still be made if providers omit or deliberately work around the technology. 27

CDSS also improve patient safety through reminder systems for other medical events, and not just those that are medication related. Among numerous examples, a CDSS for blood glucose measurement i n the ICU was able to decrease the number of hypoglycemia events. 28 This CDSS automatically prompted nurses to take a glucose measurement according to a local glucose monitoring protocol, which specified how often measurements should be done according to specific patient demographics and previous glucose levels/trends. 28

Overall, CDSS targeting patient safety through CPOE and other systems have generally been successful in reducing prescribing and dosing errors, contraindications through automated warnings, drug-event monitoring and more. 29 Patient safety can be considered a secondary objective (or requirement) of almost all types of CDSS, no matter the primary purpose for their implementation.

Clinical management

Studies have shown CDSS can increase adherence to clinical guidelines. 30 This is significant because traditional clinical guidelines and care pathways have been shown to be difficult to implement in practice with low clinician adherance. 31 , 32 The assumption that practitioners will read, internalize, and implement new guidelines has not held true. 33 However, the rules implicitly encoded in guidelines can be literally encoded into CDSS. Such CDSS can take a variety of forms, from standardized order sets for a targeted case, alerts to a specific protocol for the patients it pertains to, reminders for testing, etc. Furthermore, CDSS can assist with managing patients on research/treatment protocols, 34 tracking and placing orders, follow-up for referrals, as well as ensuring preventative care. 35

CDSS can also alert clinicians to reach out to patients who have not followed management plans, or are due for follow-up, and help identify patients eligible for research based on specific criteria. 36 A CDSS designed and implemented at Cleveland Clinic provides a point-of-care alert to physicians when a patient’s record matches clinical trial criteria. 37 The alert prompts the user to complete a form which establishes eligibility and consent-to-contact, forwards the patient’s chart to the study coordinator, and prints a clinical trial patient information sheet.

Cost containment

CDSS can be cost-effective for health systems, through clinical interventions, 38 decreasing inpatient length-of-stay, CPOE-integrated systems suggesting cheaper medication alternatives, 39 or reducing test duplication. A CPOE-rule was implemented in a pediatric cardiovascular intensive care unit (ICU) that limited the scheduling of blood count, chemistry and coagulation panels to a 24-h interval. 40 This reduced laboratory resource utilization with a projected cost savings of $717,538 per year, without increasing length of stay (LOS), or mortality.

CDSS can notify the user of cheaper alternatives to drugs, or conditions that insurance companies will cover. In Germany, many inpatients are switched to drugs on hospital drug formularies. After finding that 1 in 5 substitutions were incorrect, Heidelberg hospital developed a drug-switch algorithm and integrated it into their existing CPOE system. 41 The CDSS could switch 91.6% of 202 medication consultations automatically, with no errors, increasing safety, reducing workload and reducing cost for providers.

Administrative functions

CDSS provide support for clinical and diagnostic coding, ordering of procedures and tests, and patient triage. Designed algorithms can suggest a refined list of diagnostics codes to aid physicians in selecting the most suitable one(s). A CDSS was conceived to address inaccuracy of ICD-9 emergency department(ED) admission coding (ICD is International Statistical Classification of Diseases, standardized codes used to represent diseases and diagnoses). 42 The tool used an anatomographical interface (visual, interactive representation of the human body) linked to ICD codes to help ED physicians accurately find diagnostic admission codes faster.

CDSS can directly improve quality of clinical documentation. An obstetric CDSS featured an enhanced prompting system, significantly improving documentation of indications for labor induction and estimated fetal weight, compared to control hospital. 43 Documentation accuracy is important because it can directly aid clinical protocols. For example, a CDSS was implemented to ensure patients were properly vaccinated following splenectomy, to combat the increased risk of infections (including pneumococcal, Haemophilus influenzae , meningococcal, etc.) that comes with spleen removal. However, the authors found that 71% of patients with the term ‘splenectomy’ in their EHR did not have it documented on their problem list (which was what triggers the CDSS alert). 44 A supplemental CDSS was then developed to enhance problem list documentation of splenectomy, 45 and improve the utility of the original vaccination CDSS.

Diagnostics support

CDSS for clinical diagnosis are known as diagnostic decision support systems (DDSS). These systems have traditionally provided a computerized ‘consultation’ or filtering step, whereby they might be provided data/user selections, and then output a list of possible or probable diagnoses. 46 Unfortunately, DDSS have not had as much influence as other types of CDSS (yet) for reasons including negative physician perceptions and biases, poor accuracy (often due to gaps in data availability), and poor system integration requiring manual data entry. 47 , 48 The latter is improving with better EHR-integration and standardized vocabulary like Snomed Clinical Terms.

A good example of an effective DDSS is one which was created by Kunhimangalam et al. 49 for diagnosis of peripheral neuropathy using fuzzy logic. Through 24 input fields which include symptoms and diagnostic test outputs, they achieved 93% accuracy compared to experts at identifying motor, sensory, mixed neuropathies, or normal cases. While this has great utility, especially in countries with less access to established clinical experts, there is also a desire for systems that can supplement specialist diagnostics. DXplain is an electronic reference based DDSS that provides probable diagnosis based on clinical manifestations. 50 In a randomized control trial involving 87 family medicine residents, those randomized to use the system showed significantly higher accuracy (84% vs. 74%) on a validated diagnosis test involving 30 clinical cases. 50

Given the known incidence of diagnostic errors, particularly in primary care, 51 there is a lot of hope for CDSS and IT solutions to bring improvements to diagnosis. 52 We are now seeing diagnostic systems being developed with non-knowledge-based techniques like machine learning, which may pave the way for more accurate diagnosis. The Babylon AI powered Triage and Diagnostic System in the U.K. is a good example of the potential, but also of the work that still has to be done before these systems are ready for primetime. 53 , 54

Diagnostics support: imaging

Knowledge-based imaging CDSS are typically used for image ordering, where CDSS can aid radiologists in selecting the most appropriate test to run, providing reminders of best practice guidelines, or alerting contraindications to contrast, for example. 55 An interventional CDS for image ordering at Virginia Mason Medical Center was shown to substantially decrease the utilization rate of lumbar MRI for low back pain, head MRI for headache, and sinus CT for sinusitis. 56 The CDS required a series of questions to be answered by providers prior to image ordering (POC), to verify appropriateness. Importantly, if an image was denied, an alternative was suggested by the system. Another commercialized example is RadWise®, which guides clinicians to the most relevant imaging order by analyzing patient symptoms and matching them with a large database of diagnoses, while also providing appropriate use recommendations at the point of care. 57

There is great interest in non-knowledge based CDS for enhanced imaging and precision radiology (‘radiomics’). 58 , 59 With images accounting for increasing amounts of medical data, but requiring extensive manual interpretation, providers need technologies to aid them in extracting, visualizing, and interpreting. 60 AI technologies are proving capable of providing insights into data beyond what humans can. 61 To do so, these technologies make use of advanced pixel recognition and image classification algorithms, most prominently: deep learning (DL). 62 IBM Watson Health, DeepMind, Google, and other companies are at the forefront, developing products for use in tumor detection, 63 medical imaging interpretation, 64 diabetic retinopathy diagnosis, 65 Alzheimer’s diagnosis through multimodal feature learning, 62 and countless more. IBM Watson’s ‘Eyes of Watson’, has been able to combine image recognition of a brain scan with text recognition of case descriptions to provide comprehensive decision support (or what IBM describes as a ‘cognitive assistant’). 60

Several projects have been able to demonstrate performance that is disputably ‘on par’ with human experts. 65 , 66 , 67 , 68 For example, Google’s team trained a deep convolutional neural network (CNN) to detect diabetic retinopathy (blood vessel damage in the eye) from a dataset of 130,000 retinal images with a very high sensitivity and specificity. 65 The algorithms performance was on par with US board certified ophthalmologists. Another study just recently published by the Stanford group demonstrated a CNN for detecting arrhythmias on electrocardiogram that exceeded the accuracy (F1 and sensitivity with matched specificity) of the average cardiologist on all rhythm classes. 68 With the current rate of progress, some experts controversially speculate that in 15–20 years, the majority of diagnostic imaging interpretation will be done (or at least pre-processed) by computers. 69 For the time being however, we should think of these early systems as an addition or augmentation to a clinician’s available toolset.

Diagnostics support: laboratory and pathology

Another subset of diagnostics where CDSS can be useful is laboratory testing and interpretation. Alerts and reminders for abnormal lab results are simple and ubiquitous in EHR systems. CDSS can also extend the utility of lab-based tests for the purpose of avoiding riskier or more invasive diagnostics. In Hepatitis B and C testing, liver biopsies are considered the gold standard for diagnosis, while non-invasive lab tests are not accurate enough to be accepted. However; AI models are being developed that combine multiple tests (serum markers, imaging, and gene tests) to produce much greater accuracy. 70 There is also application for CDSS as an interpretation tool where a test’s reference ranges are highly personalized, for example age, sex, or disease subtypes. 71

Pathology reports are crucial as decision points for many other medical specialties. Some CDSS can be used for automated tumor grading. This was done for urinary bladder tumor grading and estimating recurrence, with up to 93% accuracy. 72 The same has been done for brain tumor classification and grading. 73 There are many other examples including computerized ECG analysis, automated arterial blood gas interpretation, protein electrophoresis reports, and CDSS for blood cell counting. 46

Patient-facing decision support

With the advent of the ‘Personal Health Record’ (PHR), we are seeing CDS functionality integrated, similar to EHRs, with the patient as the end user or ‘manager’ of the data. This is a great step towards patient-focused care, and CDS-supported PHRs are the ideal tool to implement shared decision-making between patient and provider, specifically because CDSS can remove a ‘lack of information’ as a barrier to a patient’s participation in their own care. 74 PHRs are frequently designed as an extension of commercial EHR software, or as standalone web-based or mobile-based applications. 75 When connected to EHRs, PHRs can have a two way relationship, whereby information entered directly by the patient can be available to their providers, and also information in the EHR can be transmitted to the PHR for patients to view. 76

One of the earliest PHRs, the “Patient Gateway”, was simply a dashboard for patients to view medications and labs, and communicate with their physicians. 77 This has expanded and some systems now allow patients to modify their own record of care, effecting the EHR data as well. 78 Another example is Vanderbilt University’s MyHealthAtVanderbilt, a PHR fully integrated into the institutional EHR. In addition to disease-targeted delivery of patient educational materials, they incorporated a Flu Tool for patients with flu-like symptoms to decide the level of care they need and then help them seek treatment. 79 Symptom tracking is a useful and common feature of PHRs, but the variety of collected data is virtually limitless, from allergies to insurance coverage to prescription and medication information. 80 Furthermore, PHRs and other patient monitoring applications can be designed to collect information from health devices and other wearables, to create actionable insights for providers. An excellent example exists in diabetes care. Many systems are already in use, 81 but one in particular pioneered by the Stanford School of Medicine uses a wearable glucose monitor which transmits data to an Apple device (HealthKit). 82 Apple has made HealthKit interoperable with the Epic EHR and Epic PHR, “MyChart”. This successfully allows providers to monitor glucose trends in their patients in between visits, and contact them through MyChart for follow up or urgent recommendations. The pilot study demonstrated improved provider workflow, communication with patients, and ultimately quality of care. 82 Various other medical fields are deploying similar systems for monitoring that combines PHR/EHR, wearable technologies, and CDSS, including but not limited to heart failure (cardiology), hypertension, sleep apnea, palliative/elder care, and more.

It is worth noting that as PHRs have become more advanced with CDSS capabilities, there has also been increasing emphasis on the design of these systems to serve shared decision making between patient and provider, and to be interactive tools to make patients more knowledgeable/involved in their own care. PHRs that only serve as a repository for health information are now seen as missing the mark, particularly by patients themselves. 75

Pitfalls of CDSS

Fragmented workflows.

CDSS can disrupt clinician workflow, especially in the case of stand-alone systems. Many early CDSS were designed as systems that required the provider to document or source information outside their typical workspace. CDSS also disrupt workflow if designed without human information processing and behaviors in mind. In response, CDSS have been designed using the ‘think-aloud’ method to model practitioners’ workflow and create a system with better usability. 83

Disrupted workflow can lead to increased cognitive effort, more time required to complete tasks, and less time face-to-face with patients. Even when CDSS are well integrated within existing information systems, there can be disconnect between face-to-face interactions and interaction with a computer workstation. Studies have found that practitioners with more experiential knowledge are less likely to use, and more likely to override CDSS. 84

Alert fatigue and inappropriate alerts

Studies have found up to 95% of CDSS alerts are inconsequential, and often times physicians disagree with or distrust alerts. 85 Other times they just do not read them. If physicians are presented with excessive/unimportant alerts, they can suffer from alert fatigue. 86

Disruptive alerts should be limited to more life-threatening or consequential contraindications, such as serious allergies. However; even allergy alerts can be incorrect, and clinicians will often verify themselves, especially if the source is another site/hospital/practitioner. 85 , 87 Medication alerts can also be specialty specific, but irrelevant when taken out of context. For example, an alert against using broad-spectrum antibiotics such as vancomycin may be inappropriate in ICU. 85 An alert against duplicate medications may be inappropriate in inflammatory bowel disease clinics, where the same class of drug can be applied through different administration routes for increased effect.

Impact on user skill

Prior to CPOE and CDSS, healthcare providers, pharmacists, and nurses were relied upon exclusively to double-check orders. CDSS can create the impression that verifying the accuracy of an order is unnecessary or automatic. 85 This is an important myth to dispel.

It is also important to consider the potential long-term effect of a CDSS on users. Over time a CDSS can exert a training effect, so that the CDSS itself may no longer be required. Coined the “carry-over effect”, it is most likely with CDSS that are educational in nature. 88 Conversely, providers may develop too much reliance or trust on a CDSS for a specific task. 89 This could be compared to using a calculator for mathematical operations over a long period of time, and then having poorer mental math skills. It is potentially problematic as the user has less independence and will be less equipped for that task should they switch to an environment without the CDSS.

CDSS may be dependent on computer literacy

Lack of technological proficiency can be hindering when engaging with a CDSS. 90 , 91 This can vary by the design details of the CDSS, but some have been found to be overly complex, relying too much on user skill. 90 , 92 Systems should aim to stay as close to the core functionality of the pre-existing system as possible. Regardless, all new systems have a learning period, and so baseline evaluations of users’ technological competence may be appropriate. Further training can then be provided to facilitate full use of CDSS capabilities, 93 or more explicit guidance incorporated into the CDSS’ recommendations themselves. 94 This information could be implemented as info buttons to be non-disruptive. 95

System and content maintenance

Maintenance of CDSS is an important but often neglected part of the CDSS life-cycle. This includes technical maintenance of systems, applications and databases that power the CDSS. Another challenge is the maintenance of knowledge-base and its rules, which must keep apace with the fast-changing nature of medical practice and clinical guidelines. Even the most advanced healthcare institutions report difficulty keeping their systems up to date as knowledge inevitably changes. 85 Order sets and the algorithmic rules behind the CDSS have been identified as particularly difficult. 85

Operational impact of poor data quality and incorrect content

EHRs and CDSSs rely on data from external, dynamic systems and this can create novel deficiencies. As an example, some CDSS modules might encourage ordering even when the hospital lacks adequate supplies. In a study by Ash et al. 85 , a number of experts indicated that at their hospital, Hemoccult tests or pneumococcal vaccine inventories run out quickly, but this is not communicated to the CDSS.

Medication and problem lists can be problematic, if not updated or used appropriately. At one site, the medication list might be a list of dispensations, which means patients may or may not be taking them(and thus must still be asked in person). 85 Other medication lists are generated from CPOE orders only, thus still requiring manual confirmation that patients are taking the medication. Systems that make it easy to distinguish these are ideal. It is also a major area where PHRs could create a solution, by collecting medication adherence data directly from patients.

In poorly designed systems, users may develop workarounds that compromise data, such as entering generic or incorrect data. 85 The knowledge base of CDSS is dependent on a centralized, large clinical data repository. Quality of data can affect quality of decision support. If data collection or input into the system is unstandardized, the data is effectively corrupted. You may design a system for use at the point-of-care, but when applied to real world environments and data, will not be utilized properly. The importance of using informational standards such as ICD, SNOMED, and others, cannot be understated.

Lack of transportability and interoperability

Despite ongoing development for the better part of three decades, CDSS (and even EHRs in general) suffer from interoperability issues. Many CDSS exist as cumbersome stand-alone systems, or exist in a system that cannot communicate effectively with other systems.

What makes transportability so difficult to achieve? Beyond programming complexities that can make integration difficult, the diversity of clinical data sources is a challenge. 96 There is a reluctance or perceived risk associated with transporting sensitive patient information. Positively, interoperability standards are continuously being developed and improved, such as Health Level 7 (HL7) and Fast Healthcare Interoperability Resources (FHIR). These are already being utilized in commercial EHR vendors. 97 Several government agencies, medical organizations and informatics bodies are actively supporting and some even mandating the use of these interoperability standards in health systems. 98 , 99 , 100

The cloud also offers a potential solution to interoperability (and other EHR ailments such as data sync, software updating, etc. 101 ). Cloud EHRs have open architecture, newer standards, and more flexible connectivity to other systems. 102 It is also a common misconception that data stored on a cloud is more vulnerable. This is not necessarily true. Web-based EHRs are required to store data in high-level storage centers with advanced encryption and other safeguards. They must comply with national data security standards including the Health Insurance Portability and Accountability Act (HIPAA) in the USA, Personal Information Protection and Electronic Documents Act (PIPEDA) in Canada, or the Data Protection Directive and General Data Protection Regulation (GDPR) in Europe, to name a few. 103 They can be just as safe (or just as vulnerable) as traditional, server-based architecture. 103 In fact, there are often fewer people who have access to unencrypted data in cloud storage centers vs. server-based records. 103

Financial challenges

Up to 74% of those with a CDSS said that financial viability remains a struggle. 104 Outset costs to set up and integrate new systems can be substantial. Ongoing costs can continue to be an issue indefinitely as new staff need to be trained to use the system, and system updates are required to keep pace with current knowledge.

Results from cost analyses of CDSS implementations are mixed, controversial, and sparse. 105 , 106 , 107 , 108 Whether an intervention is cost-effective depends on a wide range of factors, including those specific to the environment, both political and technological. 105 Cost benefit assessment in itself can be limited, with challenges such as a lack of standardized metrics. 107 This is an emerging research area and much work needs to be done to advance our understanding of the financial effects of CDSS.

CDSS have been shown to augment healthcare providers in a variety of decisions and patient care tasks, and today they actively and ubiquitously support delivery of quality care. Some applications of CDSS have more evidence behind them, especially those based on CPOE. Support for CDSS continues to mount in the age of the electronic medical record, and there are still more advances to be made including interoperability, speed and ease of deployment, and affordability. At the same time, we must stay vigilant for potential downfalls of CDSS, which range from simply not working and wasting resources, to fatiguing providers and compromising quality of patient care. Extra precautions and conscientious design must be taken when building, implementing, and maintaining CDSS. A portion of these considerations were covered in this review, but further review will be required in practice, especially as CDSS continue to evolve in complexity through advances in AI, interoperability, and new sources of data.

Osheroff, J. et al. Improving Outcomes with Clinical Decision Support: An Implementer’s Guide . (HIMSS Publishing, 2012).

Sim, I. et al. Clinical decision support systems for the practice of evidence-based medicine. J. Am. Med Inf. Assoc. Jamia. 8 , 527–534 (2001).

Article   CAS   Google Scholar  

De Dombal, F. Computers, diagnoses and patients with acute abdominal pain. Arch. Emerg. Med . 9 , 267–270 (1992).

Shortliffe, E. H. & Buchanan, B. G. A model of inexact resoning in medicine. Math. Biosci. 379 , 233–262 (1975).

Google Scholar  

Middleton, B., Sittig, D. F. & Wright, A. Clinical decision support: a 25 year retrospective and a 25 year vision. Yearb. Med. Inform. 25 (S 01), S103–S116 (2016).

Article   Google Scholar  

Dias, D. Wearable health devices—vital sign monitoring, systems and technologies. https://doi.org/10.3390/s18082414 (2018).

Berner, E. S. (Ed.). Clinical Decision Support Systems (Springer, New York, NY, 2007).

Osheroff, J., Pifer, E., Teigh, J., Sittig, D. & Jenders, R. Improving Outcomes with Clinical Decision Support: An Implementer’s Guide. (HIMS, 2005).

Deo, R. C. Machine learning in medicine. Circulation 132 , 1920–1930 (2015).

Article   PubMed   PubMed Central   Google Scholar  

HITECH Act Enforcement Interim Final Rule. https://www.hhs.gov/hipaa/for-professionals/special-topics/HITECH-act-enforcement-interim-final-rule/index.html .

Electronic medical record adoption model requirements. https://www.himssanalytics.org/emram , Accessed 25 Aug 2019 (2017).

Chang, F. & Gupta, N. Progress in electronic medical record adoption in Canada Recherche Les progrès dans l’ adoption du dossier médical électronique au Canada. Canadian Family Physician. 61 , 1076–1084 (2015).

Healthcare Information and Management Systems Society (HIMSS). Electronic Health Records: A Global Perspective, 2nd edn. https://s3.amazonaws.com/rdcms-himss/files/production/public/HIMSSorg/Content/files/Globalpt1-edited%20final.pdf (2010).

Nøhr, C. et al. Nationwide citizen access to their health data: analysing and comparing experiences in Denmark, Estonia and Australia. 1–11. https://doi.org/10.1186/s12913-017-2482-y (2017).

Omididan, Z. & Hadianfar, A. The role of clinical decision support systems in healthcare (1980-2010): a systematic review study. Jentashapir Sci.-Res Q. 2 , 125–134 (2011).

Kabane, S. M. Healthcare and the Effect of Technology: Developments, Challenges and Advancements: Developments, Challenges and Advancements. Medical Information Science Reference (2010).

Vonbach, P., Dubied, A., Krähenbühl, S. & Beer, J. H. Prevalence of drug-drug interactions at hospital entry and during hospital stay of patients in internal medicine. Eur. J. Intern. Med. 19 , 413–420 (2008).

Article   PubMed   Google Scholar  

Helmons, P. J., Suijkerbuijk, B. O., Nannan Panday, P. V. & Kosterink, J. G. W. Drug-drug interaction checking assisted by clinical decision support: a return on investment analysis. J. Am. Med. Inf. Assoc. Jamia. 22 , 764–772 (2015).

Koutkias, V. & Bouaud, J. Contributions from the 2017 Literature on Clinical Decision Support. Yearb. Med. Inf. 27 , 122–128 (2018).

Phansalkar, S. et al. High-priority drug – drug interactions for use in electronic health records. J. Am. Med. Inform. Assoc . 19 , 735–743 (2012).

Cornu, P., Phansalkar, S., Seger, D. L., Cho, I. & Pontefract, S. International Journal of Medical Informatics High-priority and low-priority drug – drug interactions in di ff erent international electronic health record systems: a comparative study. Int. J. Med. Inf. 111 , 165–171 (2018).

McEvoy, D. S. et al. Variation in high-priority drug-drug interaction alerts across institutions and electronic health records. J. Am. Med. Inf. Assoc. 24 , 331–338 (2017).

Cho, I., Lee, J., Choi, J., Hwang, H. & Bates, D. W. National rules for drug-drug interactions: are they appropriate for tertiary hospitals?. J. Korean Med. Sci. 31 , 1887–1896 (2016).

Mahoney, C. D., Berard-Collins, C. M., Coleman, R., Amaral, J. F. & Cotter, C. M. Effects of an integrated clinical information system on medication safety in a multi-hospital setting. Am. J. Health Syst. Pharm. 64 , 1969–1977 (2007).

Peris-Lopez, P., Orfila, A., Mitrokotsa, A. & van der Lubbe, J. C. A. A comprehensive RFID solution to enhance inpatient medication safety. Int. J. Med. Inf. 80 , 13–24 (2011).

Levtzion-korach, O. et al. Effect of bar-code technology on the safety of medication administration. N. Engl. J. Med. 362 , 1698–1707 (2010).

van der Veen, W. et al. Association between workarounds and medication administration errors in bar-code-assisted medication administration in hospitals. J. Am. Med. Inf. Assoc. 25 , 385–392 (2018).

Eslami, S. et al. Effects of two different levels of computerized decision support on blood glucose regulation in critically ill patients. Int J. Med. Inf. 81 , 53–60 (2012).

Jia, P., Zhang, L., Chen, J., Zhao, P. & Zhang, M. The effects of clinical decision support systems on medication safety: an overview. PLoS ONE 11 , 1–17 (2016).

Kwok, R., Dinh, M., Dinh, D. & Chu, M. Improving adherence to asthma clinical guidelines and discharge documentation from emergency departments: Implementation of a dynamic and integrated electronic decision support system. Emerg. Med. Australas. 21 , 31–37 (2009).

PubMed   Google Scholar  

Davis, D. A. & Taylor-Vaisey, A. Translating guidelines into practice: a systematic review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. Can. Med. Assoc. J. 157 , 408–416 (1997).

CAS   Google Scholar  

Michael, C., Rand, C. S., Powe, N. R., Wu, A. W. & Wilson, M. H. Why don’ t physicians follow clinical practice guidelines? a framework for improvement. Jama,. 282 , 1458–1465 (1999).

Shortliffe, T. Medical thinking: what should we do? http://www.openclinical.org/medicalThinking2006Summary2.html (2006).

Lipton, J. A. et al. Impact of an alerting clinical decision support system for glucose control on protocol compliance and glycemic control in the intensive cardiac care unit. Diabetes Technol. Ther. 13 , 343–349 (2011).

Salem, H. et al. A multicentre integration of a computer led follow up in surgical oncology is valid and safe. BJU Int . https://doi.org/10.1111/bju.14157 (2018).

Health Information Technology Foundations Module 28: Clinical Decision Support Basics. Carnegie Mellon University Open Learning Initiative. https://oli.cmu.edu/jcourse/workbook/activity/page?context=e6f7c0b180020ca600c0f4e5957d6f8c .

Embi, P. J., Jain, A., Clark, J. & Harris, C. M. Development of an electronic health record-based Clinical Trial Alert system to enhance recruitment at the point of care. AMIA Annu. Symp. Proc. 2005 , 231–235 (2005).

Calloway, S., Akilo, H. & Bierman, K. Impact of a clinical decision support system on pharmacy clinical interventions, documentation efforts, and costs. Hosp. Pharm. 48 , 744–752 (2013).

McMullin, S. T. et al. Impact of an evidence-based computerized decision support system on primary care prescription costs. Ann. Fam. Med. 2 , 494–498 (2004).

Algaze, C. A. et al. Use of a checklist and clinical decision support tool reduces laboratory use and improves cost. Pediatrics 137 , e20143019 (2016).

Pruszydlo, M. G., Walk-Fritz, S. U., Hoppe-Tichy, T., Kaltschmidt, J. & Haefeli, W. E. Development and evaluation of a computerised clinical decision support system for switching drugs at the interface between primary and tertiary care. BMC Med. Inf. Decis. Mak. 12 , 1 (2012).

Bell, C. M., Jalali, A. & Mensah, E. A decision support tool for using an ICD-10 anatomographer to address admission coding inaccuracies: a commentary. Online J. Public Health Inform. 5 , 222 (2013).

Haberman, S. et al. Effect of clinical-decision support on documentation compliance in an electronic medical record. Obstet. Gynecol. 114 , 311–317 (2009).

Turchin, A., Shubina, M. & Gandhi, T. NLP for patient safety: splenectomy and pneumovax. In Proc. AMIA 2010 Annual Symposium (2010).

McEvoy, D., Gandhi, T. K., Turchin, A. & Wright, A. Enhancing problem list documentation in electronic health records using two methods: the example of prior splenectomy. BMJ Qual Saf . https://doi.org/10.1136/bmjqs-2017-006707 (2017).

Berner E. Clinical Decision Support Systems: Theory and Practice 3rd edn. https://doi.org/10.1007/978-0-387-38319-4 (2016).

Berner, E. S. Diagnostic decision support systems: why aren’t they used more and what can we do about it? AMIA Annu. Symp. Proc . 2006 , 1167–1168 (2006).

Segal, M. M. et al. Experience with integrating diagnostic decision support software with electronic health records: benefits versus risks of information sharing. EGEMs Gener. Evid. Methods Improv. Patient Outcomes 5 , 23 (2017).

Kunhimangalam, R., Ovallath, S. & Joseph, P. K. A clinical decision support system with an integrated EMR for diagnosis of peripheral neuropathy. J. Med. Syst. 38 , 38 (2014).

Martinez-Franco, A. I. et al. Diagnostic accuracy in Family Medicine residents using a clinical decision support system (DXplain): a randomized-controlled trial. Diagn. Berl. Ger. 5 , 71–76 (2018).

Singh, H., Schiff, G. D., Graber, M. L., Onakpoya, I. & Thompson, M. J. The global burden of diagnostic errors in primary care. BMJ Qual. Saf. 26 , 484–494 (2017).

Singh, H., Meyer, A. N. D. & Thomas, E. J. The frequency of diagnostic errors in outpatient care: Estimations from three large observational studies involving US adult populations. BMJ Qual. Saf. 23 , 727–731 (2014).

Razzaki, S. et al. A comparative study of artificial intelligence and human doctors for the purpose of triage and diagnosis. arXiv preprint arXiv:1806.10698 (2018).

Fraser, H., Coiera, E. & Wong, D. Safety of patient-facing digital symptom checkers. Lancet 392 , 2263–2264 (2018).

Georgiou, A., Prgomet, M., Markewycz, A., Adams, E. & Westbrook, J. I. The impact of computerized provider order entry systems on medical-imaging services: a systematic review. J. Am. Med. Inf. Assoc. 18 , 335–340 (2011).

Blackmore, C. C., Mecklenburg, R. S. & Kaplan, G. S. Effectiveness of clinical decision support in controlling inappropriate imaging. JACR 8 , 19–25 (2019).

DSS Inc. Radiology Decision Support (RadWise®). https://www.dssinc.com/products/integrated-clinical-products/radwise-radiology-decision-support/ .

Giardino, A. et al. Role of imaging in the era of precision medicine. Acad. Radiol. 24 , 639–649 (2017).

Oakden-rayner, L. et al. Precision Radiology: predicting longevity using feature engineering and deep learning methods in a radiomics framework. Sci. Rep . https://doi.org/10.1038/s41598-017-01931-w (2017).

From Invisible to Visible: IBM Demos AI to Radiologists. https://www-03.ibm.com/press/us/en/pressrelease/51146.ws . Accessed Aug 2019 (2016).

Greenspan, H., Ginneken van, B. & Summers, R. M. Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Trans. Med. Imaging 35 , 1153–1159 (2016).

Suzuki, K. & Chen, Y. Artificial intelligence in decision support systems for diagnosis in medical imaging. https://doi.org/10.1007/978-3-319-68843-5 (2018).

IBM Watson Health - IBM Watson for Oncology. https://www.ibm.com/watson/health/oncology-and-genomics/oncology/ . Accessed 25 Aug 2019 (2018).

Lunit Inc. https://lunit.io/en/ . Accessed Aug 2019 (2018).

Gulshan, V. et al. Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA - J. Am. Med. Assoc. 316 , 2402–2410 (2016).

Esteva, A. et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature 542 , 115 (2017).

Article   CAS   PubMed   Google Scholar  

Rajpurkar, P. et al. Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning. arXiv preprint ar Xiv:1711.05225 (2017).

Hannun, A. Y. et al. FOCUS | Letters FOCUS | Letters Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network letters | FOCUS letters | FOCUS. Nat. Med . https://doi.org/10.1038/s41591-018-0268-3 (2019).

Erickson, B. J. Machine Intelligence in Medical Imaging (Society for Imaging Informatics, SIIM, 2016).

Keltch, B., Lin, Y. & Bayrak, C. Comparison of AI techniques for prediction of liver fibrosis in hepatitis patients patient facing systems. J. Med. Syst . https://doi.org/10.1007/s10916-014-0060-y (2014).

Mørkrid, L. et al. Continuous age- and sex-adjusted reference intervals of urinary markers for cerebral creatine deficiency syndromes: a novel approach to the definition of reference intervals. Clin. Chem. 61 , 760–768 (2015).

Article   PubMed   CAS   Google Scholar  

Spyridonos, P., Cavouras, D., Ravazoula, P. & Nikiforidis, G. A computer-based diagnostic and prognostic system for assessing urinary bladder tumour grade and predicting cancer recurrence. Med. Inf. Internet Med. 27 , 111–122 (2002).

Tsolaki, E. et al. Fast spectroscopic multiple analysis (FASMA) for brain tumor classification: a clinical decision support system utilizing multi-parametric 3T MR data. Int. J. Comput. Assist Radio. Surg. 10 , 1149–1166 (2015).

Davis, S., Roudsari, A., Raworth, R., Courtney, K. L. & Mackay, L. Shared decision-making using personal health record technology: a scoping review at the crossroads. J. Am. Med. Inf. Assoc. 24 , 857–866 (2017).

Fuji, K. T. et al. Standalone personal health records in the United States: meeting patient desires. Health Technol. 2 , 197–205 (2012).

Tang, P. C., Ash, J. S., Bates, D. W., Overhage, J. M. & Sands, D. Z. Personal health records: definitions, benefits, and strategies for overcoming barriers to adoption. J. Am. Med. Inf. Assoc. 13 , 121–126 (2006).

Wald, J. S. et al. A patient-controlled journal for an electronic medical record: issues and challenges. Stud. Health Technol. Inform. 107 (Pt 2), 1166–1170 (2004).

Hanauer, D. A., Preib, R., Zheng, K. & Choi, S. W. Patient-initiated electronic health record amendment requests. J. Am. Med. Inf. Assoc. 21 , 992–1000 (2014).

Rosenbloom, S. T. et al. Triaging patients at risk of influenza using a patient portal. J. Am. Med. Inf. Assoc. 19 , 549–554 (2012).

Roehrs, A., Da Costa, C. A., Da Rosa Righi, R. & De Oliveira, K. S. F. Personal health records: A systematic literature review. J. Med. Internet Res . https://doi.org/10.2196/jmir.5876 (2017).

Benhamou, P. Y. Improving diabetes management with electronic health records and patients’ health records. Diabetes Metab. 37 (Suppl. 4), S53–S56 (2011).

Kumar, R. B., Goren, N. D., Stark, D. E., Wall, D. P. & Longhurst, C. A. Automated integration of continuous glucose monitor data in the electronic health record using consumer technology. J. Am. Med. Inf. Assoc. 23 , 532–537 (2016).

Kilsdonk, E., Peute, L. W., Riezebos, R. J., Kremer, L. C. & Jaspers, M. W. M. Uncovering healthcare practitioners’ information processing using the think-aloud method: From paper-based guideline to clinical decision support system. Int. J. Med. Inf. 86 , 10–19 (2016).

Dowding, D. et al. Nurses’ use of computerised clinical decision support systems: A case site analysis. J. Clin. Nurs. 18 , 1159–1167 (2009).

Ash, J. S., Sittig, D. F., Campbell, E. M., Guappone, K. P. & Dykstra, R. H. Some unintended consequences of clinical decision support systems. AMIA Annu Symp. Proc. AMIA Symp. AMIA Symp. 2007 , 26–30 (2007).

Khalifa, M. & Zabani, I. Improving utilization of clinical decision support systems by reducing alert fatigue: Strategies and recommendations. Stud. Health Technol. Inform. 226 , 51–54 (2016).

Van Laere, S. et al. Clinical decision support systems for drug allergy checking: systematic review. https://doi.org/10.2196/jmir.8206 (2018).

Wyatt, J. & Spiegelhalter, D. Field trials of medical decision-aids: potential problems and solutions. American Medical Informatics Association. 3–7. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2247484/ (1991).

Goddard, K., Roudsari, A. & Wyatt, J. Automation bias - A hidden issue for clinical decision support system use. Stud. Health Technol. Inform. 164 , 17–22 (2011).

Devaraj, S., Sharma, S. K., Fausto, D. J., Viernes, S. & Kharrazi, H. Barriers and facilitators to clinical decision support systems adoption: a systematic review. J. Bus Adm. Res . https://doi.org/10.5430/jbar.v3n2p36 (2014).

Leslie, S. J. et al. Clinical decision support software for management of chronic heart failure: Development and evaluation. Comput Biol. Med. 36 , 495–506 (2006).

Murray, E. et al. Why is it difficult to implement e-health initiatives? A qualitative study. Implement Sci. 6 , 6 (2011).

Lai, F., Macmillan, J., Daudelin, D. H. & Kent, D. M. The potential of training to increase acceptance and use of computerized decision support systems for medical diagnosis. Hum. Factors J. Hum. Factors Erg. Soc. 48 , 95–108 (2006).

Ojeleye, L. Ensuring effective computerised clinical decision support. Prescriber 27 , 54–56 (2016).

Cook, D. A., Teixeira, M. T., Heale, B. S. E., Cimino, J. J. & Del Fiol, G. Context-sensitive decision support (infobuttons) in electronic health records: a systematic review. J. Am. Med Inf. Assoc. 24 , 460–468 (2017).

Sujansky, W. Heterogeneous database integration in biomedicine. J. Biomed. Inform. 34 , 285–298 (2001).

Index - FHIR v3.0.1. (2018). https://www.hl7.org/fhir/index.html . Accessed July 2019.

Katehakis, D. G. Towards the Development of a National eHealth Interoperability Framework to Address Public Health Challenges in Greece. Proceedings of the First International Workshop on Semantic Web Technologies for Health Data Management, SWH@ISWC. 2164 , 1–9 (2018).

EHRIntelligence. 5 Ways States Mandate Health Information Exchange. https://ehrintelligence.com/news/5-ways-states-mandate-health-information-exchange . Accessed Aug 2019 (2015).

European Commission Report. Commission Recommendation on a European Electronic Health Record Exchange Format (C(2019)800) of 6 February 2019. https://ec.europa.eu/digital-single-market/en/news/recommendation-european-electronic-health-record-exchange-form (2019).

Bresnick, J. HealthITAnalytics. Interoperability, Low Costs Make Cloud-Based EHRs a Favorite. https://healthitanalytics.com/news/interoperability-low-costs-make-cloud-based-ehrs-a-favorite . Accessed July 2019 (2015).

Fernández-Cardeñosa, G., De La Torre-Díez, I., López-Coronado, M. & Rodrigues, J. J. P. C. Analysis of cloud-based solutions on EHRs systems in different scenarios. J. Med. Syst. 36 , 3777–3782 (2012).

Rodrigues, J. J. P. C., de la Torre, I., Fernández, G. & López-Coronado, M. Analysis of the security and privacy requirements of cloud-based electronic health records systems. J. Med. Internet Res . 15 , e186 (2013).

Kabachinski, J. A look at clinical decision support systems. Biomed. Instrum. Technol. 47 , 432–434 (2013).

O’Reilly, D., Tarride, J.-E., Goeree, R., Lokker, C. & McKibbon, K. A. The economics of health information technology in medication management: a systematic review of economic evaluations. J. Am. Med. Inf. Assoc. 19 , 423–438 (2012).

Bright, T. J. et al. Effect of clinical decision-support systems: a systematic review. Ann. Intern Med. 157 , 29–43 (2012).

Jacob, V. et al. Cost and economic benefit of clinical decision support systems (CDSS) for cardiovascular disease prevention: a community guide systematic review. J. Am. Med. Inf. Assoc. Jamia. 24 , 669–676 (2017).

Main, C. et al. Computerised decision support systems in order communication for diagnostic, screening or monitoring test ordering: systematic reviews of the effects and cost-effectiveness of systems. Health Technol. Assess. 14 , 1–227 (2010).

CAS   PubMed   Google Scholar  

Scheepers-Hoeks, A. M., Grouls, R. J., Neef, C. & Korsten, H. H. Strategy for implementation and first results of advanced clinical decision support in hospital pharmacy practice. Stud. Health Technol. Inform . 148 , 142–148 (2009).

Edlin, R., McCabe, C., Hulme, C., Hall, P. & Wright, J. Cost effectiveness modelling for health technology assessment. https://doi.org/10.1007/978-3-319-15744-3 (2015).

Vermeulen, K. M. et al. Cost-effectiveness of an electronic medication ordering system (CPOE/CDSS) in hospitalized patients. Int J. Med, Inf. 83 , 572–580 (2014).

Okumura, L. M., Veroneze, I., Burgardt, C. I. & Fragoso, M. F. Effects of a computerized provider order entry and a clinical decision support system to improve cefazolin use in surgical prophylaxis: a cost saving analysis. Pharm. Pract. 14 , 1–7 (2016).

Osheroff, J. A. et al. A roadmap for national action on clinical decision support. J. Am. Med. Inf. Assoc. 14 , 141–145 (2007).

Greenes, R. A. Clinical Decision Support 2nd edn. The Road to Broad Adoption https://doi.org/10.1016/B978-0-12-398476-0.00035-X (2014).

Bonney, W. Impacts and risks of adopting clinical decision support systems. In: Efficient Decision Support Systems - Practice and Challenges in Biomedical Related Domain. IntechOpen https://doi.org/10.5772/711 (2011).

Bates, D. W. et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J. Am. Med. Inf. Assoc. 10 , 523–530 (2003).

Kawamoto, K. Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features critical to success. Bmj 330 , 765–0 (2005).

Health Level Seven International - Homepage. http://www.hl7.org/ . Accessed 29 Aug 2019 (2018).

IHTSDO. History of SNOMED CT. Ihtsdo. http://www.ihtsdo.org/snomed-ct/what-is-snomed-ct/history-of-snomed-ct (2015).

Marco-Ruiz, L. & Bellika, J. G. Semantic Interoperability in Clinical Decision Support Systems: A Systematic Review. Stud. Health Technol. Inf. 216 , 958 (2015).

Angraal, S., Krumholz, H. M. & Schulz, W. L. Blockchain technology: applications in health care. Circ. Cardiovasc. Qual. Outcomes https://doi.org/10.1161/CIRCOUTCOMES.117.003800 (2017).

Ivan, D. Moving toward a blockchain-based method for the secure storage of patient records. ONC/NIST Use of Blockchain for Healthcare and Research Workshop . Gaithersburg, Maryland, United States: ONC/NIST (2016).

Eichner, J. & Das, M. Challenges and Barriers to Clinical Decision Support (CDS) Design and Implementation Experienced in the Agency for Healthcare Research and Quality CDS Demonstrations. Agency Healthc Res. Qual. Website . 29. https://healthit.ahrq.gov/sites/default/files/docs/page/CDS_challenges_and_barriers.pdf (2010).

Khalifa, M. Clinical decision support: strategies for success. Procedia Comput. Sci. 37 , 422–427 (2014).

Sittig, D. F. Electronic Health Records: Challenges in Design and Implementation . (CRC Press, 2014).

Harper, B. D. & Norman, K. L. Improving User Satisfaction: The Questionnaire for User Interaction Satisfaction Version 5.5. Proceedings of the 1st Annual Mid-Atlantic Human Factors Conference, (pp. 224–228), Virginia Beach, VA. (1993).

Lewis, J. R. The system usability scale: past, present, and future. Int J. Hum.-Comput. Interact. 34 , 577–590 (2018).

Lewis, J. R. IBM computer usability satisfaction questionnaires: psychometric evaluation and instructions for use. Int J. Hum.-Comput. Interact. 7 , 57–78 (1995).

Lewis, J. R. Measuring perceived usability: the CSUQ, SUS, and UMUX. Int J. Hum.-Comput Interact. 34 , 1148–1156 (2018).

Download references

Acknowledgements

The author’s (R.T.S.) work was supported by a Canada Institute of Health (CIHR) Research Graduate Scholarship (CGS-M). Thank you to Nathan Stern for discussion and initial review of the manuscript.

Author information

Authors and affiliations.

Department of Medicine, Division of Gastroenterology, University of Alberta, Edmonton, Canada

Reed T. Sutton, Daniel C. Baumgart, Daniel C. Sadowski, Richard N. Fedorak & Karen I. Kroeker

Chief Medical Information Office, Alberta Health Services, Edmonton, Canada

David Pincock

You can also search for this author in PubMed   Google Scholar

Contributions

R.T.S. reviewed the literature, drafted and revised the manuscript. D.P., D.C.S., D.C.B., K.I.K., and R.N.F. provided critical input, discussion, and revision of the manuscript. All authors approved the final submitted version of the manuscript.

Corresponding author

Correspondence to Karen I. Kroeker .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Sutton, R.T., Pincock, D., Baumgart, D.C. et al. An overview of clinical decision support systems: benefits, risks, and strategies for success. npj Digit. Med. 3 , 17 (2020). https://doi.org/10.1038/s41746-020-0221-y

Download citation

Received : 21 February 2019

Accepted : 19 December 2019

Published : 06 February 2020

DOI : https://doi.org/10.1038/s41746-020-0221-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

The association between population health management tools and clinician burnout in the united states va primary care patient-centered medical home.

• Lucinda Leung
• Susan E. Stockdale

BMC Primary Care (2024)

Health professionals’ acceptance of mobile-based clinical guideline application in a resource-limited setting: using a modified UTAUT model

• Addisalem Workie Demsash
• Mulugeta Hayelom Kalayou
• Agmasie Damtew Walle

BMC Medical Education (2024)

Title and abstract screening for literature reviews using large language models: an exploratory study in the biomedical domain

• Fabio Dennstädt
• Johannes Zink
• Nikola Cihoric

Systematic Reviews (2024)

Improve the efficiency and accuracy of ophthalmologists’ clinical decision-making based on AI technology

• Yingxuan Guo
• Changke Huang

BMC Medical Informatics and Decision Making (2024)

Exploring the role of professional identity in the implementation of clinical decision support systems—a narrative review

• Sophia Ackerhans
• Thomas Huynh
• Carsten Schultz

Implementation Science (2024)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

• Open access
• Published: 30 November 2022

Decision support system for handling control decisions and decision-maker related to supply chain

• Dimah Hussein Alahmadi 1 &
• Arwa A. Jamjoom   ORCID: orcid.org/0000-0002-5062-8942 1  

Journal of Big Data volume  9 , Article number:  114 ( 2022 ) Cite this article

5579 Accesses

4 Citations

1 Altmetric

Metrics details

The present study presents a knowledge-based DSS framework for supporting the decision-maker and handling control decisions related to supply chains.

Two binary variables were adopted for denoting at which time point a given task \(i\) starts and finishes. The scheduling issues are represented through the ontological model and appropriately interpreted using the Java environment. Regarding primary data, digital form of administration using google form platform took precedence over physical administration.

The findings might not be exact replication of the findings from previous studies that are limited to the influence of information and material flow on the performance of supply chain as there are concerns of what factors constitute information and material flow that need to be identified and considered. However, with the finding of associating factors of information and material flow may need to consider this in managing the flow and the supply chain. Associating factors such as information quality, information visibility, material cost, fund shortage and so on, play a role in information and material flow and the decisions made in an organization.

Conclusions

Factors associating with information and material flow need to be considered in decision making as well, as the cost in any of the elements affects the flow and this would impede supply chain performance of the organization.

Introduction

Information technology supports business activities in fast-flowing information and prompts changes of customer preference era [ 1 ]. The recent trend causes a paradigm shift in the production process, which consequently impacts the supply chain flows, with the risk of inefficiency and overexploitation from upstream to downstream. According to Carter and Rogers [ 2 ], sustainable supply chains focus on environmental, social, and economic aspects. In this regard, a decision support system (DSS) is emerged from the recent trend and is further competent in supporting the significant issues in the supply chains. Gorry and Scott Morton [ 3 ] have initially proposed DSS, and it has been broadly utilized in several realms. DSS aims to support decision-makers in aiding and enhancing their decisions about the process and the consequence of their business functions, which are in the representation of guidance for selecting the optimum sets of options in elevating the profit, customer satisfaction, and efficiency concerning the product.

Extant literature has focused on the use of DSS to support business-related procedures. For instance, these areas include the oil industry, fisheries, marine affairs, environmental sciences, transportation, tourism, and the health sector [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 ]. This pattern indicates a very innovative application of DSS linked with numerous tools and classifications with other approaches for supporting the decision-making process [ 12 ]. DSS further experiences criticism when existing and potential consumers do not always take benefit of DSS in supporting their decision-making, either because of the DSS structure or knowledge and awareness. The consumer repeatedly and often utilizes the DSS when the usefulness and easiness are there. Therefore, DSS has to be customized based on problems and activities [ 13 ]. DSS has adopted data mining, business intelligence, statistical analysis, and data warehouse. The existing function of DSS is not merely restricted to the database system but also an expert system that aids decision-makers in solving the issues.

The efficacy of DSS is further reliant on the characteristics and construction, specifically in the supply chains, where it requires availability and information for transferring supply and demand between each stratum from downstream to upstream, enabling DSS to aid the decision-maker in the supply chains. There is a broad horizon for developing each decision support system in the supply chains since previous literature has grown significantly in the supply chain realm. Therefore, a knowledge-based DSS framework has been presented in this paper for supporting the decision-maker and handling control decisions related to supply chains.

Literature review

Organizations have always had the development of efficient supply chain systems as their focus. According to Attaran and Attaran [ 14 ], collaboration in the supply chain practice is becoming the crux of successful and long-lasting management of business operations. They posited that the inclination to produce quality goods and services is driving up the supply chain cost, affecting the supplier’s financial performance. Therefore, the significance of the supply chain to the production and eventual financial strength of the business makes it a top issue for the organization’s management [ 15 ]. For an organization to run a sustainable and successful supply chain, though, management has to ensure collaborative planning due to its effect on the movement of goods and services. As indicated by Cassivi [ 16 ], collaborative planning, forecasting, and replenishment (CPFR) is the bone of the business process that strengthens the management of the supply chain in an organization.

A successful supply chain management goes a long way in benefiting a company in a competition. Some of these benefits are evident in improving the cost of production, distribution, inventory, and the flexibility associated with the ‘production of goods and services, and improvement in market share [ 17 ] and customer relations [ 18 ]. Simatupang and Sridharan [ 19 ] posited that flexibility is vital in measuring the success of the supply chain in the organization, and it’s one of the benefits of having effective supply chain management. According to Hsu [ 20 ], the benefits of supply chain management can be either tangible or intangible. Tangible benefits are exemplified in the cost and reduction of inventory and its effective management. The time saved when the inventory is delivered quickly, while the intangible information accuracy, consistency, flow, service quality, and response time [ 21 ]. These, however, can only be realized when there is an integration of various functions and stakeholders in the organization.

According to Stevenson and Spring [ 22 ], correct and instantaneous information flow in the supply chain is equally as vital to the business as material flow. “An information-enriched supply chain would have a single customer entity connected to every scheduling process, showing order information flowing to all links, while for a non-enriched supply chain, the customer entity might connect only to the final scheduling link, leaving the remainder of the supply chain hidden from the customer” [ 23 ].

Sharing of information is essential due to its reflection of teamwork within the supply chain of an organization [ 24 ]. Simatupang and Sridharan [ 25 ] refer to information sharing as “the ability to see private data in a partner’s systems and monitor the progress of products as they pass through each process in the supply chain; the activity includes monitoring (data capturing), processing, and dissemination of customer data, end-to-end inventory status and locations, order status, costs-related data, and performance status”. Simatupang and Sridharan [ 19 ] believe that sharing information among supply chain partners allows short time order fulfillment within the order cycle times due to the shared information. Supply chain partners sharing of information generates supply chain information flow management that aids effective decision making among partners. Li et al. [ 24 ] stated that information flow is categorized based on the area of operation it is generated and needed. Therefore, the categories of information flow include production plan, inventory, order state, demand forecasting, and sales [ 26 ]. In their paper, Koh, Saad, and Arunachalam [ 27 ] stated that information flow is needed to support the management of activities like procurement of raw materials, schedule for production, and physical distribution system [ 28 ].

For information flow to be complete, two-way communication needs to be conducted, which involves contents, medium/channel, and systems [ 29 ]. The content is the actual information to be passed; the medium/channel is the pathway for the information, while the system allows the management of both information and the channel. According to Kembro and Selviaridis [ 30 ], information in supply chain information is sharable into three levels within the organization: strategic, tactical, and operational. At each of these levels, different types of information are communicated while various associated advantages and hindrances are encountered in sharing the information in the supply chain. Hsu et al. [ 31 ] also separated information shared in the organization into diverse levels, which are: strategic information (e.g., long-term objective, marketing, and customer information) and tactical (e.g., purchasing, operations scheduling, and logistics).

Information exchanged can also be classified into managerial and transactional information. Transactional refers to information required for an organization to conduct procurement or supplies. This class of information is related to payment order, receipt, inventory, transportation, and delivery. This information class relates to the technology needed for operations, quality, costs, and profitability.

From the perspective of supply chain management, information flow and its management are critical activities of the leaders in an organization. The flow of information in the supply chain is bi-directional. This is because other forms of activities, including materials and money flows, are activated by the movement of information to achieve the set objective. This means that material and money flows effective management is positively related to effective management of information flow. Therefore, the huge interest in these flows in literature and supply chain practice is understandable. Several supply chain practitioners have identified the significance of materials flow management as a critical strategic achievement issue [ 32 ].

Kuck et al. [ 33 ] suggested a data-driven simulation-based optimization approach for dynamic manufacturing system control. The study devised a method for rescheduling production in response to changing circumstances, taking into account aspects that may confuse, such as the simultaneous delivery of many orders. Ersoz et al. [ 34 ] attempted to bridge the scheduling theory and practice gap. They tailored their planning operations to the real-time information supplied by the process control and control systems. The dynamic structure of the production environment is quickly sensed in the offered procedure, and the schedule is modified in response to the changing conditions. In manufacturing, the traceability of the parts improved. Furthermore, needless waiting or downtimes were reduced.

Xiong et al. [ 35 ] suggested a simulation-based methodology for deciding dispatching rules in a dynamic scheduling issue with task release times and extended technical priority limitations. The proposed approach decreased total delay and the number of late tasks. Zhang et al. [ 36 ] conducted a literature study on job-shop scheduling issues and explored fresh viewpoints in the context of Industry 4.0. They added that under Industry 4.0, scheduling issues are addressed using new methodologies and approaches. The findings suggest that scheduling research should focus on smart distributed scheduling modeling and optimization. According to their assessment, this may be accomplished through two methods: combining old techniques and presenting a new way, as well as proposing new algorithms for smart distributed scheduling (Fig.  1 ).

Supply chain process

In their study, Rossit et al. [ 37 ] introduced the notion of intelligent manufacturing that arose with Industry 4.0. They have addressed the subject of smart scheduling, which they feel has a significant role in today’s product knowledge. They created the notion of tolerance scheduling in a dynamic environment to avoid production rescheduling. Likewise, Tao et al. [ 38 ] investigated contemporary advancements in production systems and smart manufacturing technologies, as well as Industry 4.0 models and stated that dynamic scheduling is one of the significant research undertaken in the literature in the context of Industry 4.0.

Jiang et al. [ 39 ] investigated the topic of energy-efficient job-shop scheduling to reduce the total cost of energy use and finishing time. However, the problem at hand was deemed NP-Hard. As a result, they created an enhanced whale optimization technique to address this issue. They improved the whale optimization method by using dispatching rules, nonlinear convergence factors, and mutation operations. They ran simulations to demonstrate the algorithm’s usefulness. According to the findings of simulations, the algorithm delivered benefits in terms of efficiency. Ortiz et al. [ 40 ] investigated a flexible job-shop problem and offered a novel methodology for solving it. They created a novel algorithm that reduces average tardiness and discovered better solutions than the existing dispatching rules. They developed a real-world production-scheduling challenge and an effective solution for solving it.

Ding and Jiang [ 41 ] examined the impact of IoT technologies in an industrial setting. They claim that while IoT has boosted production data, these data are sometimes discontinuous, uncorrelated, and challenging to use. As a result, they devised a strategy for using priceless data. They developed an RFID-based production data analysis system for production control in IoT-enabled smart businesses. Leusin et al. [ 42 ] developed a multi-agent system in a cyber-physical approach to handling the dynamic job-shop scheduling problem. The suggested system included self-configuring characteristics in the manufacturing process. This was accomplished through the usage of agents and IoT. In the shop, real-time data was used to make more informed decisions. A real-world case study was used to test the concept. The benefits of utilizing dynamic data and IoT in industrial applications are explored (Fig.  2 ).

Supply chain planning using DSS

Following a review of the relevant literature, it was discovered that numerous models were created to establish a decision support system for continuous process improvement based on IoT-enabled data analytics. There is a need to improve the decision support system for dynamic settings that can function with various dispatching criteria. The main goal is to improve the efficiency of production management and the job-shop.

Methodology

Knowledge-based decision support systems and model management systems have used tools like artificial intelligence to provide smarter decision-maker support. In addition, a model is presented toward closing the gap between analytical and transactional models, which are utilized in the organizational and technical aspects. It further implements different hierarchical levels throughout the enterprise structure, making information quality available. In particular, different and multiple decision insights might be sufficient throughout the decision-making task, which increases the speed response of the decision-support system.

This study tackles process control decisions associated with coordination and procedural control. Thereby, the integration of control sequence steps in the equipment modules is dealt with along with the transition between control recipes. Such decisions are mentioned in the control aspects, which receive data from the scheduling point, and offer data with the actual phase. Afterward, the information flow procedures were presented at this decision level and their association with other decision levels. The decisions are associated with the integration of the control recipe. The batch operation is managed through coordination control using the control recipe scheme in the ontological model. MATLAB managed data, and the JAVA environment offers the control function from the scheduling function with the relevant data. Consequently, it is an iterative process for decision-making, which encompasses the scheduling and control levels.

Moreover, the continuous-time STN illustration relies on explaining a common time grid that is variable and authentic for all shared resources. The model implies that all tasks commencing at a predefined time guarantee the authenticity of the material balances. Two binary variables were adopted to denote when a given task starts and finishes. The scheduling issues are represented through the ontological model and appropriately interpreted using the Java environment. GAMS implemented the formulation, whereas MILP solver was utilized to execute this formulation as the implied issues were lineal.

Concept integration

The DSS framework is intended to aid in the decision-making process. Many alternatives exist, particularly those linked with Industry 4.0 technologies; nonetheless, researchers’ objective is to build a more robust system to reduce the risk of human mistakes, particularly during the data entry phase. However, the critical components of DSS and technology employed are data management, communication, the user (decision-maker), and the simulation model. The Knowledge-Based Engineering (KBE) technique was used to create the framework, which is ideal for lean products. Furthermore, knowledge in DSS and Industry 4.0 indicates the use of data management tools via IoT in this idea where human aspects (users/decision-makers) continue in developing innovation to boost productivity (Prasad, 2014) (Fig.  3 ). In conjunction, LM principles functioned as a bridge.

Concept of DSS design based on KBE (Source: Prasad, 2014)

Developing framework

Combining physical and virtual processes results in smart manufacturing (Godfrey, 2002). IoT uses the internet networking idea to collect data from sensors. The data is collected using a barcode sensor. They are then installed in the database before being sent to the server through the network. The MySQL command is provided to compile and code the data to match the simulation required input. These step procedures give the core of an Industry 4.0-compliant networking system. Simulation-based Knowledge-Based Modeling (KBM) allows users to forecast and produce reliable results based on simulation data (Prasad and Rogers, 2005). Figure  4 illustrates the enhancement of proposed system in the context of Industry 4.0.

The proposed framework

Data requirement

The data needed are specified to take decisions at the scheduling level. The phases consist of the usability to schedule predefined in this ontological model regarding all cases. In particular, the scheduling function needs the information regarding capacity, demand, due date, product stage-unit, quantities in/out, processing time, stage-process, time horizon, and unit availability. The Java code was programmed to generate the input files to schedule optimization tools.

The investigation is made for academic purposes only and not for the organizations’ promotion or human resources appraisal. The questionnaire was e-mailed to the respondents along with a consent letter which doubles as a letter of introduction of the research and what is expected of the participants. Participants were assured of their anonymity and the confidentiality of their responses. The participants were also assured that no harm would befall them on their participation in the study. The participants were informed to fill and submit the form online only if they were interested in participating in the study. Otherwise, they were imploring to ignore the mail. The responses from the participant were downloaded after a couple of weeks of initiation, and the result was analyzed. The study also searched online, via the Google search engine, for articles related to information and material flow in organizations published in reputable journals, reviewed them for collation of data on associating factors of information and material flow. The result of the review formed the basis of identifying factors related to information and material flow for the study.

Data administration

Data collected from the company staff were subjected to analysis using SPSS v20.0. Descriptive statistics, factor analysis, correlation, and regression tools were used to analyze the data. Factor analysis was done to illustrate the strength of items or associating factors of information and material flow and supply chain performance in their groups. The correlation was used to test the relationships between information and material flow and supply chain performance. On the other hand, regression analysis was deployed to understand the predictability of information and material flow of supply chain performance and which of the independent variables has a stronger contribution to supply chain performance.

The descriptive analysis explains the percentage distribution of the respondents on the characteristics of the demographic variables (Table 1 ).

The result of factor analysis of the data collated from the sampled organizations is presented. Factor analysis is the calculation and explanation of the strength of individual items concerning the group that forms the whole of the factor being investigated. Tables 2 and 3 present the strength of things that make up information and material flow and supply chain performance within organizations.

The primary data quantitatively obtained from the sampled organization was used for the analysis due to its validity in evaluating the extent of information and material flow and level of supply chain performance in the organizations. Correlation analysis explains the relationships between information and material flow and organizations’ supply chains. On the other hand, regression analysis illustrates the strength of the contribution of information and material flow on supply chain performance to explicitly ascertain the causal association in line with the third objective of the study. The correlation and regression analysis results are presented in Tables 4 and 5 .

Table 4 presents the correlation matrix analysis showing the inter-correlation between information flow, material flow, and supply chain performance. First, the table shows the mean of Information flow to be (M = 19.53; SD = 1.98); Material flow (M = 27.06; SD = 2.61); Supply chain performance (M = 38.40; SD = 2.96). The relationship between the variables shows that there are significant positive relationships between information flow and supply chain performance (r = 0.237*). Also, the result in the table shows that material flow is significant in its positive relationship with supply chain performance (r = 0.411**). This result implies that the less hampered the flow of information and material experienced, the higher the level of supply chain performance. To get a clearer picture of each of the information and material flow and their strength of prediction supply chain performance, the data is further subjected to regression analysis, and the result is presented in Table 5 .

Table 5 presents a regression result showing the relationship between information and material flow and supply chain performance. Recall that the elements of information flow considered in the study are: information quality, information accuracy, information adequacy, credibility, information timeliness, and visibility. The elements that makeup material flow include material price fluctuation, imperfect sorting, delivery delay, poor planning, fund shortage, non-alignment specification, unnecessary paperwork, and in-house logistic problem. The result showed that material flow represents the strongest factor in predicting supply chain performance. It had a significant positive relationship and accounted for about 52.3% of the variance of supply chain performance (Beta = 0.523; t = 5.648; pv = 0.000). Information flow also had a significant positive relationship, but account less for about 42.3% of supply chain performance (Beta = 0.423; t = 4.534; pv = 0.000). Collectively, information and material flow had a significant positive relationship of 54.0% with supply chain performance of organizations (R2 = 0.540; Fcal = 31.236; pv = 0.000); although there is a significant difference in the degree of contribution of the individual factors to supply chain performance as pointed out by the Fcal.

Data mining techniques can process the data present in dynamic databases to determine the problems faced in production, generate rules to control output, improve product quality, and develop automation based on work intelligence. A better understanding of production systems may aid researchers in developing sophisticated DSS and decision-makers in making better judgments. Controlling the production environment with real-time data may also assist researchers in modeling the system without making any assumptions. This technique can bridge the gap between theory and experience, resulting in more realistic solutions.

Furthermore, real job processing times might be validated using job data from the past. This study would aid in predicting the system’s behavior under various scenarios, any subsystem could simply integrate into this system. Combining artificial intelligence or machine learning-based subsystems might form the basis of future smart factories. In this regard, this study indicates the benefits of such subsystem integrations, such that even technology integration in a real job-shop may take considerable time and effort.

The findings might not be an exact replication of the results from previous studies that are limited to the influence of information and material flow on the performance of the supply chain as there are concerns of what factors constitute information and material flow that need to be identified considered. However, with the finding of associating information and material flow may need to consider this in managing the flow and the supply chain. Associating factors such as information quality, information visibility, material cost, fund shortage, and so on, play a role in information and material flow and the decisions made in an organization, which in turn impact supply chain performance, as cost of material, for example, significantly impact the flow of materials. Factors associated with information and material flow need to be considered in decision making as well, as the cost in any of the elements affects the flow, which would impede the organization's supply chain performance. As a supply chain involves a network of supply chain partners, organizations need to ensure that information quality, information visibility, material price, funding, and so on are managed effectively through a centralized policy as they provide significant insights into the health of the supply chain.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Dellino G, Laudadio T, Mari R, Mastronardi N, Meloni C. A reliable decision support system for fresh food supply chain management. Int J Prod Res. 2018;56(4):1458–85.

Article   MATH   Google Scholar  

Carter CR, Rogers DS. A framework of sustainable supply chain management: moving toward new theory. Int J Phys Distrib Logist Manag. 2008. https://doi.org/10.1108/09600030810882816 .

Article   Google Scholar  

Gorry GA, Morton MSS. A framework for management information systems. Sloan Manag Rev 1971;13:55–70.

Hemmat M, Ayatollahi H, Maleki M, Saghafi F. Health information technologies in Iran: opportunities for development. Med J Islam Repub Iran. 2019;33:103.

Google Scholar  

Belgium S, Gorunescu F. How can intelligent decision support systems help the medical research? In: Belciug S, Gorunescu F, editors. Intelligent decision support systems—a journey to smarter healthcare. Berlin: Springer; 2020. p. 71–102.

Rico N, Díaz I, Villar JR, de la Cal E. Intelligent decision support to determine the best sensory guardrail locations. Neurocomputing. 2019;18(354):41–8.

El Abdallaoui HE, El Fazziki A, Ennaji FZ, Sadgal M. Decision support system for the analysis of traffic accident Big Data. In: 2018 14th International Conference on Signal-Image Technology and Internet-Based Systems (SITIS). IEEE. 2018. pp. 514–521.

Tan WJ, Yang CF, Château PA, Lee MT, Chang YC. Integrated coastal zone management for sustainable tourism using a decision support system based on system dynamics: a case study of Cijin, Kaohsiung. Taiwan Ocean Coast Manag. 2018;1(153):131–9.

Sholahuddin A, Shadriyah Y. Analysis of students’ process skills and chemistry learning outcomes. In: The 5th South East Asia Development Research Conference (SEA-DR) International Conference. Amsterdam: Altlantis Press; 2017. p. 364–370.

Garmendia E, Gamboa G, Franco J, Garmendia JM, Liria P, Olazabal M. Social multi-criteria evaluation as a decision support tool for integrated coastal zone management. Ocean Coast Manag. 2010;53(7):385–403.

Amir-Heidari P, Arneborg L, Lindgren JF, Lindhe A, Rosén L, Raie M, Axell L, Hassellöv IM. A state-of-the-art model for spatial and stochastic oil spill risk assessment: a case study of oil spill from a shipwreck. Environ Int. 2019;1(126):309–20.

Govindan K, Mina H, Alavi B. A decision support system for demand management in healthcare supply chains considering the epidemic outbreaks: A case study of coronavirus disease 2019 (COVID-19). Transp Res Part E: Logist Transp Rev. 2020;1(138):101967.

Koh SL, Genovese A, Acquaye AA, Barratt P, Rana N, Kuylenstierna J, Gibbs D. Decarbonising product supply chains: design and development of an integrated evidence-based decision support system–the supply chain environmental analysis tool (SCEnAT). Int J Prod Res. 2013;51(7):2092–109.

Attaran M, Attaran S. Collaborative supply chain management: the most promising practice for building efficient and sustainable supply chains. Bus Proc Manag J. 2007. https://doi.org/10.1108/14637150710752308 .

Salmon K. Análisis estratégico de la cadena de suministros fibras-textil-vestido (México).

Cassivi L. Collaboration planning in a supply chain. Supply Chain Manag: Int J. 2006. https://doi.org/10.1108/13598540610662158 .

Shepherd C, Günter H. Measuring supply chain performance: current research and future directions. In: Fransoo JC, Waefler T, Wilson JR, editors. Behavioral operations in planning and scheduling. Berlin: Springer; 2010. p. 105–21.

Chapter   Google Scholar  

Ferguson BR. Implementing supply chain management. Prod Invent Manag J. 2000;41(2):64.

Simatupang TM, Sridharan R. A benchmarking scheme for supply chain collaboration. Benchmarking: Int J. 2004. https://doi.org/10.1108/14635770410520285 .

Hsu PH, Wee HM. Horizontal suppliers coordination with uncertain suppliers deliveries. Int J Oper Res. 2005;2(2):17–30.

MATH   Google Scholar  

Alam A, Bagchi PK. Supply chain capability as a determinant of FDI. Multinatl Bus Rev. 2011. https://doi.org/10.1108/15253831111172658 .

Stevenson M, Spring M. Flexibility from a supply chain perspective: definition and review. Int J Oper Prod Manag. 2007. https://doi.org/10.1108/01443570710756956 .

Hull B. A structure for supply-chain information flows and its application to the Alaskan crude oil supply chain. Logist Inf Manag. 2002. https://doi.org/10.1108/09576050210412639 .

Li G, Yan H, Wang S, Xia Y. Comparative analysis on value of information sharing in supply chains. Supply Chain Manag: Int J. 2005. https://doi.org/10.1108/13598540510578360 .

Simatupang TM, Sridharan R. The collaborative supply chain. Int J Logist Manag. 2002;13(1):15–30.

Lee H, Whang S. Decentralized multi-echelon supply chains: incentives and information. Manage Sci. 1999;45(5):633–40.

Koh SL, Saad S, Arunachalam S. Competing in the 21st century supply chain through supply chain management and enterprise resource planning integration. Int J Phys Distrib Logist Manag. 2006. https://doi.org/10.1108/09600030610677401 .

Bovet D, Martha J. Supply chain hidden profits. Mercer Management Consulting. 2003. https://books.google.com/books?id=dJsFh-LIKEkC&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false . Accessed 8 Aug 2003.

Durugbo C, Tiwari A, Alcock JR. Managing integrated information flow for delivery reliability. Ind Manag Data Syst. 2014. https://doi.org/10.1108/IMDS-10-2013-0430 .

Kembro J, Selviaridis K. Exploring information sharing in the extended supply chain: an interdependence perspective. Supply Chain Manag: Int J. 2015. https://doi.org/10.1108/SCM-07-2014-0252 .

Hsu CW, Hu AH. Green supply chain management in the electronic industry. Int J Environ Sci Technol. 2008;5(2):205–16.

Tummala VR, Phillips CL, Johnson M. Assessing supply chain management success factors: a case study. Supply Chain Manag: Int J. 2006. https://doi.org/10.1108/13598540610652573 .

Kück M, Ehm J, Hildebrandt T, Freitag M, Frazzon EM. Potential of data-driven simulation-based optimization for adaptive scheduling and control of dynamic manufacturing systems. In: 2016 Winter Simulation Conference (WSC). IEEE. 2016. p. 2820–2831.

Ersöz S, Türker AK, Aktepe A. Üretim Süreçlerinin Optimizasyonunda RFID Teknolojisi ve Uzman Sistem Temelli Tümle¸sik Yapının ERP Sistemine Entegrasyonu ve FNSS Savunma Sistemleri A. ¸S.’de Uygulanması; San-Tez Project Report; Ankara, Turkey, 2016. https://adnanaktepe.com/projeler/ . Accessed 14 Mar 2019.

Xiong H, Fan H, Jiang G, Li G. A simulation-based study of dispatching rules in a dynamic job shop scheduling problem with batch release and extended technical precedence constraints. Eur J Oper Res. 2017;257(1):13–24.

Article   MathSciNet   MATH   Google Scholar  

Zhang J, Ding G, Zou Y, Qin S, Fu J. Review of job shop scheduling research and its new perspectives under Industry 4.0. J Intell Manuf. 2019;30(4):1809–30.

Rossit DA, Tohmé F, Frutos M. Industry 4.0: smart scheduling. Int J Prod Res. 2019;57(12):3802–13.

Tao F, Qi Q, Liu A, Kusiak A. Data-driven smart manufacturing. J Manuf Syst. 2018;1(48):157–69.

Jiang T, Zhang C, Zhu H, Gu J, Deng G. Energy-efficient scheduling for a job shop using an improved whale optimization algorithm. Mathematics. 2018;6(11):220.

Ortíz MA, Betancourt LE, Negrete KP, De Felice F, Petrillo A. Dispatching algorithm for production programming of flexible job-shop systems in the smart factory industry. Ann Oper Res. 2018;264(1):409–33.

Ding K, Jiang P. RFID-based production data analysis in an IoT-enabled smart job-shop. IEEE/CAA J Autom Sin. 2017;5(1):128–38.

Leusin ME, Frazzon EM, Uriona Maldonado M, Kück M, Freitag M. Solving the job-shop scheduling problem in the industry 4.0 era. Technologies. 2018;6(4):107.

Download references

Acknowledgements

The author is highly thankful for all the associated personnel who contributed to the completion of this study. This research holds no conflict of interest.

This research received no external funding.

Author information

Authors and affiliations.

Information System Department, King Abdulaziz University, Jeddah, Saudi Arabia

Dimah Hussein Alahmadi & Arwa A. Jamjoom

You can also search for this author in PubMed   Google Scholar

Contributions

DHA: analysis & drafting. AAJ: conception, administration, analysis, editing & approval. Both authors read and approved the final manuscript.

Corresponding author

Correspondence to Arwa A. Jamjoom .

Ethics declarations

Ethics approval and consent to participate.

Not applicable.

Consent for publication

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Alahmadi, D.H., Jamjoom, A.A. Decision support system for handling control decisions and decision-maker related to supply chain. J Big Data 9 , 114 (2022). https://doi.org/10.1186/s40537-022-00653-9

Download citation

Received : 25 October 2021

Accepted : 19 September 2022

Published : 30 November 2022

DOI : https://doi.org/10.1186/s40537-022-00653-9

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Decision support system
• Continuous process improvement
• Data analytics
• Supply chains

Reference management. Clean and simple.

How to format your references using the Decision Support Systems citation style

This is a short guide how to format citations and the bibliography in a manuscript for Decision Support Systems . For a complete guide how to prepare your manuscript refer to the journal's instructions to authors.

• Using reference management software

Typically you don't format your citations and bibliography by hand. The easiest way is to use a reference manager:

• Journal articles

Those examples are references to articles in scholarly journals and how they are supposed to appear in your bibliography.

Not all journals organize their published articles in volumes and issues, so these fields are optional. Some electronic journals do not provide a page range, but instead list an article identifier. In a case like this it's safe to use the article identifier instead of the page range.

• Books and book chapters

Here are examples of references for authored and edited books as well as book chapters.

Sometimes references to web sites should appear directly in the text rather than in the bibliography. Refer to the Instructions to authors for Decision Support Systems.

This example shows the general structure used for government reports, technical reports, and scientific reports. If you can't locate the report number then it might be better to cite the report as a book. For reports it is usually not individual people that are credited as authors, but a governmental department or agency like "U. S. Food and Drug Administration" or "National Cancer Institute".

• Theses and dissertations

Theses including Ph.D. dissertations, Master's theses or Bachelor theses follow the basic format outlined below.

• News paper articles

Unlike scholarly journals, news papers do not usually have a volume and issue number. Instead, the full date and page number is required for a correct reference.

• In-text citations

References should be cited in the text by sequential numbers in square brackets :

• About the journal

• Other styles
• Cancer Immunology, Immunotherapy
• Frontiers in Evolutionary Neuroscience

Pros and Cons of Clinical Decision Support Systems (CDSS) Essay

• To find inspiration for your paper and overcome writer’s block
• As a source of information (ensure proper referencing)
• As a template for you assignment

Kilsdonk, E., Peute, L. W., & Jaspers, M.W. M. (2017). Factors influencing implementation success of guideline-based clinical decision support systems: A systematic review and gaps analysis. International Journal of Medical Informatics , 98 , 56-64. doi: 10.1016/j.ijmedinf.2016.12.001

• Cardiotocography Training Development Project in Saudi Arabia
• SLP Assignment: Leadership and Cultural Competence of Healthcare Professionals
• Concept of Decision Support (DS)
• Healthcare Fraud and Kickbacks
• Healthcare Fraud and the Welfare of the Healthcare Consumers
• How Understanding of Family Influenced Assessment
• Evaluating the Treatment of a Sexually Assaulted Child
• Workplace Violence Prevention Bill for Healthcare Workers
• The Introduction of New Systems Into the Work of Medical Care
• Health Inequalities in the Developing Countries
• Chicago (A-D)
• Chicago (N-B)

IvyPanda. (2023, February 12). Pros and Cons of Clinical Decision Support Systems (CDSS). https://ivypanda.com/essays/pros-and-cons-of-clinical-decision-support-systems-cdss/

"Pros and Cons of Clinical Decision Support Systems (CDSS)." IvyPanda , 12 Feb. 2023, ivypanda.com/essays/pros-and-cons-of-clinical-decision-support-systems-cdss/.

IvyPanda . (2023) 'Pros and Cons of Clinical Decision Support Systems (CDSS)'. 12 February.

IvyPanda . 2023. "Pros and Cons of Clinical Decision Support Systems (CDSS)." February 12, 2023. https://ivypanda.com/essays/pros-and-cons-of-clinical-decision-support-systems-cdss/.

1. IvyPanda . "Pros and Cons of Clinical Decision Support Systems (CDSS)." February 12, 2023. https://ivypanda.com/essays/pros-and-cons-of-clinical-decision-support-systems-cdss/.

Bibliography

IvyPanda . "Pros and Cons of Clinical Decision Support Systems (CDSS)." February 12, 2023. https://ivypanda.com/essays/pros-and-cons-of-clinical-decision-support-systems-cdss/.

We use cookies to enhance our website for you. Proceed if you agree to this policy or learn more about it.

• Essay Database >
• Essay Examples >
• Essays Topics >
• Essay on Business

Example Of Decision Support System Essay

Type of paper: Essay

Topic: Business , Information , Entrepreneurship , Organization , Decision , Corporation , Commerce , Company

Published: 01/30/2020

ORDER PAPER LIKE THIS

Decision support system

Decision making is one of the activities that take place within organizations on a daily basis, and it is one of the major roles that corporation’s manager or the management team is involved in, and entrusted to carry out for the corporation. As a result, therefore, there have been many opinions and approaches towards this activity, since it carries a lot of weight and relevant for corporations. It is also necessary to note beforehand, that decision making is one of the most essential roles for every corporation, as it involves taking major steps towards the achievement of the corporation’s set goals and objectives. In this case, therefore, many corporations, especially in middle level and large scale corporations and organizations, have adopted the decision support system (Information Builders, 2012). The decision support system, otherwise referred to as the DSS, is a computerized info system, which supports a company or organization’s decision making steps, process and activities. As it has already been highlighted before, there are several activities and processes that take place in every organization, irrespective of its size, with the help, assistance and monitoring of the corporation’s managers. These activities include planning, management, decision making and operations, and each of these activities take place at their own distinct levels. Some of these activities and processes in the organizations might be inconsistent changing rapidly, constantly and unexpectedly. There are various reasons that normally lead to these unexpected changes, and these are some of the reasons why corporations have adopted the system, and some of the reasons for these changes include the current inconsistent changes I the world economy, the consistent and rapid developments and changes in the information and computer technology field (which leads to changes in the corporation’s activities and setting) as well as other internal changes that might affect the corporation, leading to changes of the same. The role of the Decision Support System is top therefore, monitor and effectively manage all these changes, in order to maintain the corporations’ activities at an optimum (Turban, 2010). One of the most essential things to note before adopting a decision support system for an organization is to note the fact that the support system can operate in a number of ways, whereby it can be fully computerized, manually (human) operated, or in other cases, depending with the operations of the corporation and the activities that take place at the organization, the decision system can be a combination of both the human and the computerized systems. This is one of the major advantages of adopting the system, because it serves to fit the corporation in accordance with the organization of the corporation (Turban & Aronson, 2004). It is necessary to note the fact that there have been other decision making systems that various companies, organizations and corporations have adopted to assist them in decision making, as well as keeping a trail on the changes taking place in the corporation, and these include expert system, knowledge base system, as well as group decision support systems. However, the decision support system has been found to be one of the most effective and adoptable support systems, based on several properties. Firstly, it is necessary to note the fact that the system is flexible, and versatile in various ways. For example, the support system adopts the computerized and manual systems, as well as a combination of both. This is necessary because it makes it applicable in the small scale, middle level and even the large scale corporations (Turban, 2010). In this case, therefore, no size of corporation will be affected by the level of development or function ability of the corporation, as the corporation can adopt the system irrespective of its level of information and computer technology advancement in its strategies and operations. At the same time, the decision support system is an interactive software, and its main role is to assist the decision makers to collect and compile only the useful information from raw data collected in the field, for example during research. At the same time, personal knowledge, skill and opinion, information from documents as well as other sources are considered raw data, and the decision support systems assists in analyzing the same, in order to come up with only the useful information to be used in any project or purpose, thereby adopting only the information that is verified as relevant. This is one of the reasons why many corporations, especially in the twenty first century, have adopted this system (Turban & Aronson, 2004). In conclusion, some of the unique roles that the system can perform and that give it an upper hand compare to the other three earlier mentioned systems include accessing and assessing all the information assets in a corporation’s database, comparing information (quantitative information) between one period and the next (for example on weekly basis), projecting revenue figures and guiding the corporation’s management team on the possible consequences arising from various decisions, based on past experiences and outcomes from similar or related decisions (Information Builders, 2012). This system has, therefore, come in handy and helpful to organizations, and has been adopted since it assists in not only planning, but also time saving and forecasting on the future trends of the corporations.

Information Builders. (2012). “Decision Support System: DSS.” Information Builders. Retrieved from http://www.informationbuilders.com/decision-support-systems-dss on April 3rd, 2013. Turban, E. (2010). Decision Support and Business Intelligence Systems (9th Edition). New York: McGraw-Hill Press, pp. 233-43. Turban, E., & Aronson, J. (2004). Decision Support Systems and Intelligent Systems (7th Edition). New York: McGraw-Hill Press, pp. 146-48.

Cite this page

Share with friends using:

Removal Request

Finished papers: 1920

This paper is created by writer with

ID 260040626

If you want your paper to be:

Well-researched, fact-checked, and accurate

Original, fresh, based on current data

Eloquently written and immaculately formatted

275 words = 1 page double-spaced

Get your papers done by pros!

Other Pages

Confirmation hearing essays, copra essays, phonation essays, collarbone essays, seta essays, conservation of mass essays, minuet essays, comparative literature essays, mental health critical thinking examples, sociology interviews creative writing, free essay on the harlem renaissance through cullen and mckay, the cultures essay, report on john q student, example of hydrographs of flooding on the indus river pakistan 2010 case study, essay on andrew jacksons war with the second bank of the united states, free research paper on the stuxnet, pop art manifesto research paper sample, example of case study on jenkins v american express financial corporation, jane doro report, similarity research paper examples, max wertheimers contribution to gestalt theory and how it is used today research paper example, what approach would the nurse use to elicit the cooperation of a school age child case study, research paper on impact of accessibility of assault rifles, review of attendance monitoring procedure creative writing example, example of essay on your e book is reading you, affirmative action a discussion critical thinking examples, example of quiet time creative writing, good purpose and objectives of the study essay example, free critique to andersons democratic egalitarianism essay top quality sample to follow, key trends essay samples, pfizer overview research paper samples, inspiring movie review about the role of digital diplomacy in an interconnected world, essay on defining the approach, global energy a sample essay for inspiration mimicking, crimes against women research paper to use for practical writing help, good essay on critical reflection 2, fuel efficient stove final report free sample report to follow, free capstone journal essay top quality sample to follow, sample case study on ronworking case study, free global expansion for small businesses research paper sample, expertly crafted essay on guilty or not guilty does it matter, the risk factors for the development of stroke after heart surgery literature review example.

Password recovery email has been sent to [email protected]

Use your new password to log in

You are not register!

By clicking Register, you agree to our Terms of Service and that you have read our Privacy Policy .

Now you can download documents directly to your device!

Check your email! An email with your password has already been sent to you! Now you can download documents directly to your device.

or Use the QR code to Save this Paper to Your Phone

The sample is NOT original!

Short on a deadline?

Don't waste time. Get help with 11% off using code - GETWOWED

No, thanks! I'm fine with missing my deadline

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• BMJ Health Care Inform
• v.31(1); 2024
• PMC11344512

What characteristics of clinical decision support system implementations lead to adoption for regular use? A scoping review

1 Sport and Exercise Medicine, Queen Mary University, London, UK

Dylan Morrissey

William marsh.

2 Electronic Engineering and Computer Science, Queen Mary University, London, UK

Associated Data

All data relevant to the study are included in the article or uploaded as online supplemental information.

Introduction

Digital healthcare innovation has yielded many prototype clinical decision support (CDS) systems, however, few are fully adopted into practice, despite successful research outcomes. We aimed to explore the characteristics of implementations in clinical practice to inform future innovation.

Web of Science, Trip Database, PubMed, NHS Digital and the BMA website were searched for examples of CDS systems in May 2022 and updated in June 2023. Papers were included if they reported on a CDS giving pathway advice to a clinician, adopted into regular clinical practice and had sufficient published information for analysis. Examples were excluded if they were only used in a research setting or intended for patients. Articles found in citation searches were assessed alongside a detailed hand search of the grey literature to gather all available information, including commercial information. Examples were excluded if there was insufficient information for analysis. The normalisation process theory (NPT) framework informed analysis.

22 implemented CDS projects were included, with 53 related publications or sources of information (40 peer-reviewed publications and 13 alternative sources). NPT framework analysis indicated organisational support was paramount to successful adoption of CDS. Ensuring that workflows were optimised for patient care alongside iterative, mixed-methods implementation was key to engaging clinicians.

Extensive searches revealed few examples of CDS available for analysis, highlighting the implementation gap between research and healthcare innovation. Lessons from included projects include the need for organisational support, an underpinning mixed-methods implementation strategy and an iterative approach to address clinician feedback.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Many studies report success in developing clinical decision support systems, but the vast majority do not make it beyond research.

WHAT THIS STUDY ADDS

• This study summarises the common characteristics of those clinical decision support implementations that have made it into routine use in clinical practice.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• The implications of this study are a guide to researchers on how to maximise the success of their clinical decision support system.

Clinical decision support (CDS) systems have received significant focus in recent research and development activity. The rise of digital innovation, moving from paper flow charts or questionnaires to online apps and machine-learning software has given exciting opportunities to CDS developers. A Google Scholar search for ‘CDS’ in the last year yielded over 100 000 hits, in medical specialisms from acute kidney injury, 1 pneumonia, 2 hypertension 3 and cancer. 4 CDS has been described as promising to improve diagnostic accuracy. 5 Despite copious research, few CDS systems have been adopted into routine care. 6 7

Many CDS research studies focus on quantitative accuracy, improving decision-making by increasing the number of ‘correct’ diagnoses made 5 or decision-making consistency 8 which is often variable, irrespective of decision-maker expertise. 9 CDS study reports often recommend progression to clinical trials once acceptable accuracy is achieved 10 11 with no apparent consideration for implementation factors such as workflow, roles and responsibilities, and clinician engagement. Mair et al 12 found that implementation studies often considered organisational issues, but not wider issues influencing usage. Implementation models have been proposed to address qualitative aspects of new technology, 13 14 but there are no reviews of these models’ impact on adoption. This absence limits the justification of costly mixed-methods implementation studies.

NPT 15 is a framework for implementation research that has been successfully used to reveal implementation factors concerning digital health innovations for electronic medical records 16 and telehealth consultations. 17 The NPT describes going from ‘novel to normal’, essential to achieving adoption and sustainability of digital innovations. Greenhalgh and Abimbola 7 produced a comprehensive framework to address Non-adoption, Scale-up, Spread and Sustainability (NASSS) of health technologies and was considered for this study. However, the absence of specific details about each individual innovation made it impossible to use fully. The technology acceptance model 18 19 and the critical success factor framework 20 were potential alternatives considered but were not specific to implementation research or shared domains with the NPT so were rejected after detailed analysis ( online supplemental appendix 1 ).

We aimed to identify CDS system features associated with adoption into routine use using the NPT. Key objectives were to summarise common strategies leading to successful implementation and normalised use and identify strategies to inform future implementations

We followed the JBI scoping reviews guidance 21 22 and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA) checklist, 23 with the protocol published on the Open Science Framework. 24

Eligibility criteria

CDS systems were included if they met clear definitions ( table 1 ). We focused on clinician-facing CDS systems due to their particular workflow challenges and excluded developmental or patient-facing systems.

We searched Web of Science, OpenGrey, 25 the BMA 26 and Trip databases 27 in June 2022 with updates in April 2024. CDS systems and websites were suggested by expert colleagues and were searched in Google and Google Scholar for related publications or reports. Once a CDS was included, a further search of names and authors was conducted to deepen the analysis. See online supplemental appendix 2 for details.

Quality assessment

Included CDS projects were assessed against the Standards for Reporting Implementation Studies (STARI), 28 which assesses interventions separately from implementation, recognising the benefits of studying and reporting both in healthcare settings. See online supplemental appendix 3 for details.

Analysis framework

The NPT framework 15 was used to deductively analyse the implementation of included CDS systems, 15 as it highlights issues critical to successful implementation and integration into routine work. It is recommended for use in designing complex interventions 29 and commonly used as a qualitative data analysis framework. 30

The NPT framework is split into four domains: Coherence, cognitive participation, collective action and reflexive monitoring ( figure 1 ). The four domains are further subdivided within the NPT framework using questions to be answered during the analysis. 15

Data charting process and data items

The subquestions within the four NPT domains were answered during full-text review and data extraction, from each CDS’s associated publications, or designated ‘not reported’ where not described in the publications. Quotes to substantiate answers are given in the framework analysis ( online supplemental appendix 4 ).

Synthesis of results

Results were analysed quantitatively by calculating the percentage of positive and not reported answers by NPT domain and by project. The NPT was used as the qualitative content analysis framework.

Search results are presented in the PRISMA diagram ( figure 2 ).

Data were extracted in Covidence 31 before exporting. The projects retained for final analysis represent 22 individual CDS tools and 54 study records, as some CDS projects had multiple implementation reports ( table 2 ). All published outputs were analysed to yield a complete review. Full references for table 2 are contained in online supplemental file 5 .

CDSclinical decision supportCDSScomputerised decision support systemEHRelectronic health recordICUintensive care unitiLFTIntelligent liver function testingPICUpaediatric ICU

Of the 22 CDS tools, 18 had evidence of over half the STARI checklist domains. Many publications associated with the projects focused exclusively on either implementation or intervention. See online supplemental appendix 2 for full scoring.

The questions in each NPT domain answered from the published information for each CDS ( online supplemental appendix 4 ) were colour-coded with red for a ‘no’ or negative response; amber for a ‘maybe’, neutral or ambivalent response and green for a ‘yes’ or positive response ( table 3 ). If there was insufficient information, this was deemed ‘not reported’ and colour-coded grey ( figure 3 ). See online supplemental appendix 4 for full definitions of terms and synthesis explanation.

NPTnormalisation process theory

‘Is there a clear understanding of the new service?’ scored well with 64% positive ‘Yes’ reports ( figure 3 ). Publications provide explanations that are clear to the reader, however, there is limited reporting of the explanation given to clinicians involved. Frequently, the CDS replicate published guidelines.

‘Do individuals have a shared understanding of the aims, objectives and benefits?’ was variable. It appeared that mandating use to circumvent this aspect has been used in some cases (F, Q, V, Z) . In others, the CDS was ‘communicated’ to staff ( figure 4 ) but there was no evidence of understanding its purpose (A, H, UU). Training and targeted advertising appear to have been successful in many cases. Where training was used, the emphasis was on bespoke training by profession or clinical role.

Only 36% of studies said ‘yes’ in response to ‘Do individuals have a clear understanding of their task or responsibility?’, and several studies reported ‘no’ that is, that during implementation individuals explicitly expressed a lack of understanding (A, F, MM). Understanding role-specific tasks and responsibilities was problematic at the outset of most of the implementation projects that reported on this ( figure 4 ).

‘Do individuals understand the value, benefits and importance of the service?’ was variably reported. Where uptake of the CDS was low, implementers in some cases mandated use to achieve uptake. Learning from inappropriate requests was mentioned as a benefit among multiple CDS projects. Where clinicians were aware of improvements in costs, efficiency or compliance with guidelines, this was well received (A, B, S,KK).

Cognitive participation

There were 45% ‘yes’ responses to ‘Do individuals buy into the idea of the service?’ ( figure 3 ). Five projects relied on mandated use of the CDS to achieve uptake, rather than encouraging buy-in (F, Q, V, Z, FF). There were significant barriers to engagement in many cases. Training and teamwork appeared to be facilitating factors (H,X) along with time-saving interventions embedded in the CDS pathway or workflow (JJ).

‘Can individuals sustain involvement?’ scored 59% ‘yes’ responses. Across the board, integration with existing electronic health record (EHR) was key to sustaining involvement ( Figure 5 ) (B, K, GG).

41% of CDS did not report any evidence for ‘Are key individuals willing to drive the implementation?’ Of those that mentioned key individuals, management support was key (Z, GG, MM) and having a small team of known team members to discuss and encourage usage (F, H, CC).

‘Do individuals feel it is right for them to be involved?’ was the least reported element of the framework (73% not reported). Of the five projects that reported against this question, two reported a negative response (Figure 5) (CC,OO). Only four projects contained any report on this domain (HH, MM, OO, UU). Where problems were identified by the team undertaking the implementation, it was clear they had a mandate to be involved.

Collective action

Organisational support for the CDS project was reported positively in 45% of projects ( figure 3 ). Where it was not explicitly reported, indications of organisational support included enabling significant changes to a local care pathway (F). Support came from a wide range of sources, including clinical commissioning groups in the UK (F, CC, KK), insurance companies in the USA (Z) or professional bodies such as the Swedish Pharmacy Association (M). Organisational incentives to introduce a CDS were usually cost and efficiency savings ( figure 6 ). These priorities did not necessarily align with those mentioned by the users: safety and improved patient care. Even where the CDS had no key individuals to drive the implementation (K, RR), the organisational support led to success.

‘How does the innovation affect roles and responsibility or training needs?’ was positively reported in 41% of cases. There are examples of implementation leading to positive role change within a team ( Figure 7 ). For example, a change in the pharmacists’ and microbiologists’ roles from being exclusively lab based to having a patient-facing role was reported (KK). Where the CDS replicated best practice guidelines, subsequent service audits demonstrated improvement in adherence to guidelines long term (K, KK, NN).

There were only 32% ‘yes’ responses to ‘Does the service make people’s work easier?’. Additional time burden was a perceived disadvantage (Figure 7). Some factored in incremental changes, and therefore, the burden was mitigated in response to feedback. Examples of mitigation included increasing integration with the EHR, 32 improved automation of inputs (B) or removing another burdensome task to make the CDS timesaving or time-neutral (JJ).

‘Do individuals have confidence in the new system?’ was answered ‘yes’ in 45% of projects, however, 50% was not reported. Where there was feedback from clinicians, the studies reported confidence in the new system almost universally (B, MM, OO, UU).

Reflexive monitoring

‘Do individuals try to alter the new service ?’ had 14% ‘yes’ answers ( figure 3 ). Lack of flexibility for clinical judgement was a clear negative aspect of implementations; this led to circumventing (F), ‘gaming’ (A), or ignoring the CDS (M, CC). Nevertheless, adherence to guidelines and better patient care was frequently assessed as a benefit of the CDS. Where there was an additional, unmitigated administrative burden, CDS implementations were viewed negatively (N, Q, Z, HH). However, where mitigations were introduced, participants reported positive effects on knowledge and patient care (GG, UU).

There were 18% positive responses to: ‘How do individuals appraise the effects on them and their work environment?’. Effects on clinicians and clinical environment were often not reported (50%), and when it is, it is mostly negative—time burden (Z), alert fatigue (CC) and administrative tasks were identified downsides (GG).

‘How are benefits or problems identified or measured?’ was positively reported in 55% of CDS. Many of the projects either planned a qualitative element to their studies (H, Z, CC), or an iterative approach (B, F) to implementation that allowed for feedback. This meant that problems with a CDS could be identified once in practice and subsequently solved.

Summary of evidence

Despite extensive searches, there are few publications available for analysis regarding CDS implementations adopted into routine clinical practice. Even among those included, there are extensive areas of the NPT not explicitly addressed. CDS research studies typically focus on quantitative methods, reflected in extensive categories not reported. The NPT interrogates individuals’ responses to new technology, however, quantitative studies may only explore the aggregated numerical impact of a given measure. Projects with the most content against the NPT domains were those with a mixed-methods or primarily qualitative implementation paradigm (H, N, Z, CC, HH, MM, SS, VV).

Organisational support

Organisational prioritisation of CDS implementation was one of the most reported characteristics, and endorsement is key for CDS adoption, in turn, more likely if the best practice or national guidelines are included (A, O, V, FF, GG, OO). Having a financial or operational mandate to use a CDS from an insurance company or clinical commissioning group (F, Z, CC), or key supportive individuals leading the implementation (K, M) is an adoption promotor. The ongoing use of the CDS in these cases demonstrates the importance of organisational support in ensuring success. This was demonstrated during the COVID-19 pandemic where there was impetus to allow clinical services to continue despite rules preventing face-to-face contact; senior leadership allowed significant innovation where previously there had been stagnation. 33 There were examples of mandating use when uptake was low (F, Q, V, Z, FF); it is unclear whether this was instead of or in addition to communicating importance and value. It is, therefore, difficult to draw conclusions about the impact this had. Where there was an organisational imperative to introduce the CDS, iterative approaches and staff feedback will likely have been deprioritised. This is evidenced by the two studies with the most negative reporting against the NPT (F, Z). This goes against the ISO standards for human-centred design, 34 which recommends having users at the heart of implementation. One of the projects comments on clinical autonomy being compromised by mandating the use of the CDS (F). Mandating usage is a blunt tool which achieves the organisational aim but ignores the complexity of the patient-facing role. The evidence in this review shows improved engagement when clinicians are consulted for feedback and problem-solving instead of forced implementation. This is also reflected by Greenhalgh and Abimbola, 7 recognising the adopter is not just the organisation, but all stakeholders involved in the system’s use. Golinelli et al 33 highlighted the lack of legal frameworks surrounding CDS; mandating systems and reducing clinical autonomy without legal guidance may discourage adoption.

Shared value proposition

Understanding role-specific tasks was often reported neutrally or negatively . Conversely, key commonalities were having staff who understood the value proposition of the CDS and understood their role in using it. The manifestation of lack of understanding varied. Wilkinson et al (MM) reported an explicit lack of understanding by clinicians, whereas Hart et al (F) and Macpherson et al (B) reported the CDS was not being completed correctly or at all. This highlights the key sections of the NPT which have a positive impact are a clear understanding of the new service (CDS), individuals can sustain involvement, a positive value judgement of the CDS and the ability to identify and measure problems. This is echoed in the NASSS framework 7 and has also been demonstrated in other case studies. 35 36 Understanding that more effective care is delivered using CDS facilitates buy-in. 35

Iterative, mixed-methods and the need for customisation

In studies with qualitative feedback or questionnaires, it is easier to explore how the technology has been normalised into service delivery (N, Z, CC, II, MM, UU), illustrated by having the lowest number of domains not reported. Novel health technology implementation is recognised as a complex intervention and as such requires a comprehensive approach to all aspects of system performance and usability. 37 Mixed-methods approaches 13 combined with human-centred design 34 optimise the implementation and prevent circumventing or gaming the system. 36 This illustrates the requirement for adaptability and customising of the system to meet clinicians’ needs in practice. 36 Rolling with complexity is recommended in the NASSS framework, 7 recognising the overlapping responsibilities of clinicians and organisations and accepts that there are some complexities which will not be solved by a CDS. As such an iterative implementation in response to feedback was pivotal here(B, F, MM) and is backed up in the literature. 7 13 38

Strengths and limitations

The strength of this review lies in the exclusive inclusion of projects that are being used in regular clinical practice. This gives actionable insights relevant to many healthcare environments, for future planning of CDS implementation. Previous reviews focusing on digital adoption post pandemic have rightly highlighted the drive for innovation, 33 however, the complexities of the healthcare system and reasons for previous stagnation are rapidly being realised. 39 It is possible that the insights here could also be characteristic of failed projects since we have only considered those CDS with evidence that they are being used in routine clinical care.

Quality assessment is not strictly needed with a scoping review. However, the STARI checklist 28 acknowledges several subtypes of implementation study and the need for flexibility in the approach to judging quality in implementation. Most projects scored over half marks in the assessment. Given that these are successfully adopted implementations, it could be that the quality of the implementation study(ies) was also a factor in their success. There was a distinct pattern of certain projects which had focused on the implementation as the focus of the publication, and others which had focused on the intervention ( online supplemental appendix 1 ). As such there are elements of the checklist that the authors of the studies did not intend to address.

This review details extensive searches of the grey literature and websites, in addition to the peer-reviewed literature databases. Coupled with follow-up searches for included projects, this analysis gives a thorough picture of the CDS systems used in clinical practice where the implementation documentation is available. It is possible that examples were missed, especially where there is a lack of publications related to a particular CDS. The review has nevertheless been able to identify tangible and actionable strategies for future implementations ( figure 8 ). Future work should explore patterns in adoption such as particular clinical areas, successful funding streams and the type and range of expertise that has proven successful.

Deductive analysis of the NPT domains was relatively straightforward where there were multiple qualitative and quantitative outputs from an implementation. It is possible that NPT domains may have been addressed by the implementation project but were not reported in the associated outputs. By accessing as many sources as possible, we have attempted to minimise this eventuality.

Significant numbers of CDS systems are developed yet are never adopted in routine practice. Those that are successfully adopted have a shared value proposition and enjoy organisational support. Researchers looking to implement a CDS system should underpin their work with an iterative, mixed-methods research paradigm and should consult a published quality assessment checklist such as the STARI to address both intervention and implementation within their trial.

supplementary material

Online supplemental file 1, online supplemental file 2, online supplemental file 3, online supplemental file 4, online supplemental file 5.

Funding: Author AH received funding for an Allied Health Professional Doctoral Fellowship from Barts Charity.

Provenance and peer review: Not commissioned; externally peer reviewed.

Patient consent for publication: Not applicable.

Data availability statement

Advancing scientific understanding of climate, improving society’s ability to plan and respond

Events/Webinars

FY25 ESSM NOFO Informational Webinar

FY25 NIDIS Coping with Drought Competition Post LOI Informational Webinar

Applicant Resources at a Glance

• Current Federal Funding Opportunity (FFO)
• Data Universal Numbering System (DUNS) Number
• System for Award Management (SAM)
• National Environmental Policy Act (NEPA) Guidance
• Application for Federal Assistance  SF-424 Form SF-424 Form Instructions
• Budget Information  SF-424A SF-424A Form Instructions
• Assurances  SF-424B
• Certifications Required by the Department of Commerce  CD-511
• Catalog of Federal Domestic Assistance (CFDA)
• Grants Online
• Code of Federal Regulations (CFR)
• Department of Commerce, Grants and Cooperative Agreements Manual
• Department of Commerce, Grants Management Division

Project 1.2 Pittsburgh: Integrated Modeling and Decision Support for Stormwater Managers

The overarching goal of the proposed work is to support communities in the Mid-Atlantic region in their efforts to leverage actionable climate science to make informed decisions about adapting and building resilience to climate change, with specific attention given to the needs of vulnerable groups facing risks from stormwater flooding. Our subgroup’s work will focus on integrating modeling and decision support for stormwater managers and stakeholders in Pittsburgh. Managing stormwater associated with extreme rainfall is a major challenge for the city, and this challenge is expected to increase in response to climate change; however, transforming global climate data into actionable information to support local resilience strategies remains a challenge. The proposed work will develop and apply a dual-drainage model, i.e., both surface and subsurface stormwater runoff, that will be used to evaluate future flood hazards under climate change and support the development of equity-driven, resilient green infrastructure systems in the face of deep uncertainty. The framework and methods used to support decision making under deep uncertainty with respect to extreme rainfall and green infrastructure at local and watershed scales will help guide similar efforts in other communities and regions.

Funding Opportunities News and Events

NOAA seeks proposals to advance climate research

• August 12, 2024

Modeling, Analysis, Predictions, and Projections (MAPP) Program: Early Career Award for Exceptional Research in Earth System Model Development and Application

Atmospheric Chemistry, Carbon Cycle, and Climate (AC4) Program

FY25 NIDIS Coping with Drought Competition Information Webinar

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Formation control of a multi-unmanned surface vessel system: a bibliometric analysis.

1. Introduction

2. methodology, 2.1. research framework and data source, 2.2. bibliometric method and visualization tool, 3.1. publication trend, 3.2. social structure, 3.2.1. authors, 3.2.2. organizations, 3.2.3. countries, 3.3. citation network, 3.3.1. papers, 3.3.2. journals, 3.4. keywords and terms, 3.4.1. keywords, 3.4.2. terms, 3.5. supportive techniques for a multi-usv system, 3.5.1. cooperative path planning, 3.5.2. multi-task allocation, 3.5.3. formation control, 3.5.4. ai-powered multi-usv system, 4. discussion, 4.1. current challenges, 4.1.1. communication problem, 4.1.2. system failure, 4.1.3. dynamic and uncertain environmental conditions, 4.1.4. nonlinearity and constraints, 4.2. future directions, 4.2.1. dynamic formation switching, 4.2.2. data reduction and fast transmission, 4.2.3. algorithm improvement, 4.2.4. perception fusion, 4.2.5. real-world application scenarios, 4.3. limitations, 5. conclusions, author contributions, institutional review board statement, informed consent statement, data availability statement, conflicts of interest.

• Peng, Z.; Wang, J.; Wang, D.; Han, Q.L. An Overview of Recent Advances in Coordinated Control of Multiple Autonomous Surface Vehicles. IEEE Trans. Ind. Inform. 2021 , 17 , 732–745. [ Google Scholar ] [ CrossRef ]
• Vasilijevic, A.; Nad, D.; Mandic, F.; Miskovic, N.; Vukic, Z. Coordinated Navigation of Surface and Underwater Marine Robotic Vehicles for Ocean Sampling and Environmental Monitoring. IEEE/ASME Trans. Mechatron. 2017 , 22 , 1174–1184. [ Google Scholar ] [ CrossRef ]
• Wang, N.; Sun, J.C.; Er, M.J.; Liu, Y.C. A Novel Extreme Learning Control Framework of Unmanned Surface Vehicles. IEEE Trans. Cybern. 2016 , 46 , 1106–1117. [ Google Scholar ] [ CrossRef ]
• Fan, Y.; Huang, H.; Tan, Y. Robust Adaptive Path Following Control of an Unmanned Surface Vessel Subject to Input Saturation and Uncertainties. Appl. Sci. 2019 , 9 , 1815. [ Google Scholar ] [ CrossRef ]
• Er, M.J.; Ma, C.; Liu, T.; Gong, H. Intelligent motion control of unmanned surface vehicles: A critical review. Ocean Eng. 2023 , 280 , 114562. [ Google Scholar ] [ CrossRef ]
• Zhang, H.; Huang, Y.; Qin, H.; Geng, Z. USV Search Mission Planning Methodology for Lost Target Rescue on Sea. Electronics 2023 , 12 , 4584. [ Google Scholar ] [ CrossRef ]
• Babić, A.; Ferreira, F.; Kapetanović, N.; Mišković, N.; Bibuli, M.; Bruzzone, G.; Motta, C.; Ferretti, R.; Odetti, A.; Caccia, M.; et al. Cooperative marine litter detection and environmental monitoring using heterogeneous robotic agents. In Proceedings of the OCEANS 2023-Limerick, Limerick, Ireland, 5–8 June 2023; pp. 1–6. [ Google Scholar ]
• MahmoudZadeh, S.; Abbasi, A.; Yazdani, A.; Wang, H.; Liu, Y. Uninterrupted path planning system for Multi-USV sampling mission in a cluttered ocean environment. Ocean Eng. 2022 , 254 , 111328. [ Google Scholar ] [ CrossRef ]
• Xie, J.; Zhou, R.; Luo, J.; Peng, Y.; Liu, Y.; Xie, S.; Pu, H. Hybrid Partition-Based Patrolling Scheme for Maritime Area Patrol with Multiple Cooperative Unmanned Surface Vehicles. J. Mar. Sci. Eng. 2020 , 8 , 936. [ Google Scholar ] [ CrossRef ]
• Sinisterra, A.J.; Dhanak, M.R.; Von Ellenrieder, K. Stereovision-based target tracking system for USV operations. Ocean Eng. 2017 , 133 , 197–214. [ Google Scholar ] [ CrossRef ]
• Bai, Y.; Wang, X.; Yang, Y.; Yu, J.; Zhao, Z.; Jin, X. Autonomous Decision-Making of Unmanned Surface Vehicle Operation with Multi-Source Sensing. In Proceedings of the 2020 Chinese Automation Congress (CAC), Shanghai, China, 6–8 November 2020; pp. 2570–2575. [ Google Scholar ]
• Wang, N.; Gao, Y.; Zhang, X. Data-Driven Performance-Prescribed Reinforcement Learning Control of an Unmanned Surface Vehicle. IEEE Trans. Neural Netw. Learn. Syst. 2021 , 32 , 5456–5467. [ Google Scholar ] [ CrossRef ]
• Liu, G.; Wu, J.; Wen, N.; Zhang, R. A Review on Collaborative Planning of Multiple Unmanned Surface Vehicles. In Proceedings of the 2018 Chinese Automation Congress (CAC), Xi’an, China, 30 November–2 December 2018; pp. 2158–2163. [ Google Scholar ]
• Mostafa, M.Z.; Khater, H.A.; Rizk, M.R.; Bahasan, A.M. A novel GPS/DVL/MEMS-INS smartphone sensors integrated method to enhance autonomous navigation, guidance and control system of AUSVs based on ADSF Combined Filter. Measurement 2019 , 146 , 590–605. [ Google Scholar ] [ CrossRef ]
• Yang, P.; Xue, J.; Hu, H. A Bibliometric Analysis and Overall Review of the New Technology and Development of Unmanned Surface Vessels. J. Mar. Sci. Eng. 2024 , 12 , 146. [ Google Scholar ] [ CrossRef ]
• Jovanović, I.; Perčić, M.; BahooToroody, A.; Fan, A.; Vladimir, N. Review of research progress of autonomous and unmanned shipping and identification of future research directions. J. Mar. Eng. Technol. 2024 , 23 , 82–97. [ Google Scholar ] [ CrossRef ]
• Munim, Z.H. Autonomous ships: A review, innovative applications and future maritime business models. Supply Chain. Forum Int. J. 2019 , 20 , 266–279. [ Google Scholar ] [ CrossRef ]
• Wróbel, K.; Gil, M.; Krata, P.; Olszewski, K.; Montewka, J. On the use of leading safety indicators in maritime and their feasibility for Maritime Autonomous Surface Ships. Proc. Inst. Mech. Eng. Part O J. Risk Reliab. 2021 , 237 , 314–331. [ Google Scholar ] [ CrossRef ]
• Bibuli, M.; Singh, Y.; Sharma, S.; Sutton, R.; Hatton, D.; Khan, A. A Two Layered Optimal Approach towards Cooperative Motion Planning of Unmanned Surface Vehicles in a Constrained Maritime Environment. IFAC-PapersOnLine 2018 , 51 , 378–383. [ Google Scholar ] [ CrossRef ]
• Yao, P.; Lou, Y.; Zhang, K. Multi-USV cooperative path planning by window update based self-organizing map and spectral clustering. Ocean Eng. 2023 , 275 , 114140. [ Google Scholar ] [ CrossRef ]
• Gao, S.; Peng, Z.; Liu, L.; Wang, H.; Wang, D. Coordinated target tracking by multiple unmanned surface vehicles with communication delays based on a distributed event-triggered extended state observer. Ocean Eng. 2021 , 227 , 108283. [ Google Scholar ] [ CrossRef ]
• Brecher, J.N.; Zereik, E.; Bibuli, M.; Loth, M.; Glotzbach, T. Cooperative Distributed Navigation, Guidance, and Control Systems for Autonomous Surface Vehicles with Different Physical Connections. IFAC-PapersOnLine 2021 , 54 , 208–215. [ Google Scholar ] [ CrossRef ]
• Liu, H.; Chen, G.; Tian, X. Cooperative formation control for multiple surface vessels based on barrier Lyapunov function and self-structuring neural networks. Ocean Eng. 2020 , 216 , 108163. [ Google Scholar ] [ CrossRef ]
• Huang, Y.; Wu, D.; Li, L.; Feng, N. Event-triggered cooperative path following control of multiple underactuated unmanned surface vehicles with complex unknowns and actuator saturation. Ocean Eng. 2022 , 249 , 110740. [ Google Scholar ] [ CrossRef ]
• Tan, G.; Zhuang, J.; Zou, J.; Wan, L. Multi-type task allocation for multiple heterogeneous unmanned surface vehicles (USVs) based on the self-organizing map. Appl. Ocean Res. 2022 , 126 , 103262. [ Google Scholar ] [ CrossRef ]
• Tan, G.; Sun, H.; Du, L.; Zhuang, J.; Zou, J.; Wan, L. Coordinated control of the heterogeneous unmanned surface vehicle swarm based on the distributed null-space-based behavioral approach. Ocean Eng. 2022 , 266 , 112928. [ Google Scholar ] [ CrossRef ]
• Li, X.; Gao, M.; Kang, Z.; Sun, H.; Liu, Y.; Yao, C.; Zhang, A. Collaborative search and rescue based on swarm of H-MASSs using consensus theory. Ocean Eng. 2023 , 278 , 114426. [ Google Scholar ] [ CrossRef ]
• Zhu, Y.; Bai, J.; Guo, G. Cooperative target substitution tracking control of multiple unmanned surface vehicles with substitute USVs. Ocean Eng. 2023 , 273 , 113903. [ Google Scholar ] [ CrossRef ]
• Liang, X.; Qu, X.; Hou, Y.; Li, Y.; Zhang, R. Distributed coordinated tracking control of multiple unmanned surface vehicles under complex marine environments. Ocean Eng. 2020 , 205 , 107328. [ Google Scholar ] [ CrossRef ]
• Fan, J.; Liao, Y.; Li, Y.; Jiang, Q.; Wang, L.; Jiang, W. Formation Control of Multiple Unmanned Surface Vehicles Using the Adaptive Null-Space-Based Behavioral Method. IEEE Access 2019 , 7 , 87647–87657. [ Google Scholar ] [ CrossRef ]
• van Eck, N.J.; Waltman, L. VOS: A New Method for Visualizing Similarities Between Objects. In Proceedings of the 30th Annual Conference of the German-Classification-Society, Berlin, Germany, 8–10 March 2006; pp. 299–306. [ Google Scholar ]
• Li, J.; Goerlandt, F.; Reniers, G. An overview of scientometric mapping for the safety science community: Methods, tools, and framework. Saf. Sci. 2021 , 134 , 105093. [ Google Scholar ] [ CrossRef ]
• Veiga-del-Baño, J.M.; Cámara, M.Á.; Oliva, J.; Hernández-Cegarra, A.T.; Andreo-Martínez, P.; Motas, M. Mapping of emerging contaminants in coastal waters research: A bibliometric analysis of research output during 1986–2022. Mar. Pollut. Bull. 2023 , 194 , 115366. [ Google Scholar ] [ CrossRef ]
• Demirci, S.M.E.; Elçiçek, H. Scientific awareness of marine accidents in Europe: A bibliometric and correspondence analysis. Accid. Anal. Prev. 2023 , 190 , 107166. [ Google Scholar ] [ CrossRef ]
• Di Ciaccio, F.; Troisi, S. Monitoring marine environments with Autonomous Underwater Vehicles: A bibliometric analysis. Results Eng. 2021 , 9 , 100205. [ Google Scholar ] [ CrossRef ]
• Vasconcelos, R.N.; Lima, A.T.C.; Lentini, C.A.D.; Miranda, G.V.; Mendonça, L.F.; Silva, M.A.; Cambuí, E.C.B.; Lopes, J.M.; Porsani, M.J. Oil Spill Detection and Mapping: A 50-Year Bibliometric Analysis. Remote Sens. 2020 , 12 , 3647. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Wang, D.; Chen, Z.; Hu, X.; Lan, W. Adaptive Dynamic Surface Control for Formations of Autonomous Surface Vehicles with Uncertain Dynamics. IEEE Trans. Control Syst. Technol. 2013 , 21 , 513–520. [ Google Scholar ] [ CrossRef ]
• Kuwata, Y.; Wolf, M.T.; Zarzhitsky, D.; Huntsberger, T.L. Safe Maritime Autonomous Navigation With COLREGS, Using Velocity Obstacles. IEEE J. Ocean. Eng. 2014 , 39 , 110–119. [ Google Scholar ] [ CrossRef ]
• Shojaei, K. Leader–follower formation control of underactuated autonomous marine surface vehicles with limited torque. Ocean Eng. 2015 , 105 , 196–205. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Liu, L.; Wang, J. Output-Feedback Flocking Control of Multiple Autonomous Surface Vehicles Based on Data-Driven Adaptive Extended State Observers. IEEE Trans. Cybern. 2021 , 51 , 4611–4622. [ Google Scholar ] [ CrossRef ]
• Gu, N.; Wang, D.; Peng, Z.; Liu, L. Observer-Based Finite-Time Control for Distributed Path Maneuvering of Underactuated Unmanned Surface Vehicles with Collision Avoidance and Connectivity Preservation. IEEE Trans. Syst. Man Cybern. Syst. 2021 , 51 , 5105–5115. [ Google Scholar ] [ CrossRef ]
• Almeida, J.; Silvestre, C.; Pascoal, A. Cooperative control of multiple surface vessels in the presence of ocean currents and parametric model uncertainty. Int. J. Robust Nonlinear Control 2010 , 20 , 1549–1565. [ Google Scholar ] [ CrossRef ]
• Lu, Y.; Zhang, G.; Sun, Z.; Zhang, W. Adaptive cooperative formation control of autonomous surface vessels with uncertain dynamics and external disturbances. Ocean Eng. 2018 , 167 , 36–44. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Gu, N.; Zhang, Y.; Liu, Y.; Wang, D.; Liu, L. Path-guided time-varying formation control with collision avoidance and connectivity preservation of under-actuated autonomous surface vehicles subject to unknown input gains. Ocean Eng. 2019 , 191 , 106501. [ Google Scholar ] [ CrossRef ]
• Chen, L.; Hopman, H.; Negenborn, R.R. Distributed model predictive control for vessel train formations of cooperative multi-vessel systems. Transp. Res. Part C Emerg. Technol. 2018 , 92 , 101–118. [ Google Scholar ] [ CrossRef ]
• Shojaei, K. Observer-based neural adaptive formation control of autonomous surface vessels with limited torque. Robot. Auton. Syst. 2016 , 78 , 83–96. [ Google Scholar ] [ CrossRef ]
• Wang, N.; Jin, X.; Er, M.J. A multilayer path planner for a USV under complex marine environments. Ocean Eng. 2019 , 184 , 1–10. [ Google Scholar ] [ CrossRef ]
• Zhang, B.; Liu, W.; Mao, Z.; Liu, J.; Shen, L. Cooperative and Geometric Learning Algorithm (CGLA) for path planning of UAVs with limited information. Automatica 2014 , 50 , 809–820. [ Google Scholar ] [ CrossRef ]
• Xing, B.; Yu, M.; Liu, Z.; Tan, Y.; Sun, Y.; Li, B. A Review of Path Planning for Unmanned Surface Vehicles. J. Mar. Sci. Eng. 2023 , 11 , 1556. [ Google Scholar ] [ CrossRef ]
• Yao, P.; Zhao, R.; Zhu, Q. A Hierarchical Architecture Using Biased Min-Consensus for USV Path Planning. IEEE Trans. Veh. Technol. 2020 , 69 , 9518–9527. [ Google Scholar ] [ CrossRef ]
• Savla, K.; Frazzoli, E.; Bullo, F. Traveling Salesperson Problems for the Dubins Vehicle. IEEE Trans. Autom. Control 2008 , 53 , 1378–1391. [ Google Scholar ] [ CrossRef ]
• Faigl, J. An Application of Self-Organizing Map for Multirobot Multigoal Path Planning with Minmax Objective. Comput. Intell. Neurosci. 2016 , 2016 , 2720630. [ Google Scholar ] [ CrossRef ]
• Liu, Y.; Bucknall, R. Efficient multi-task allocation and path planning for unmanned surface vehicle in support of ocean operations. Neurocomputing 2018 , 275 , 1550–1566. [ Google Scholar ] [ CrossRef ]
• Guan, W.; Luo, W.; Cui, Z. Intelligent decision-making system for multiple marine autonomous surface ships based on deep reinforcement learning. Robot. Auton. Syst. 2024 , 172 , 104587. [ Google Scholar ] [ CrossRef ]
• Jin, X.; Zhang, G.; Yang, M.; Peng, X.; Shi, B. Cooperative multi-task traversing with complex marine environment for multiple unmanned surface vehicles inspired by membrane computing. Ocean Eng. 2022 , 266 , 112586. [ Google Scholar ] [ CrossRef ]
• Liu, Y.; Song, R.; Bucknall, R.; Zhang, X. Intelligent multi-task allocation and planning for multiple unmanned surface vehicles (USVs) using self-organising maps and fast marching method. Inf. Sci. 2019 , 496 , 180–197. [ Google Scholar ] [ CrossRef ]
• Afonso, R.J.M.; Maximo, M.R.O.A.; Galvão, R.K.H. Task allocation and trajectory planning for multiple agents in the presence of obstacle and connectivity constraints with mixed-integer linear programming. Int. J. Robust Nonlinear Control 2020 , 30 , 5464–5491. [ Google Scholar ] [ CrossRef ]
• Wang, Z.; Yan, X. Multi-UAV Task Assignment Based on Quantum Genetic Algorithm. J. Phys. Conf. Ser. 2021 , 1824 , 012010. [ Google Scholar ] [ CrossRef ]
• Patel, R.; Rudnick-Cohen, E.; Azarm, S.; Otte, M.; Xu, H.; Herrmann, J.W. Decentralized Task Allocation in Multi-Agent Systems Using a Decentralized Genetic Algorithm. In Proceedings of the 2020 IEEE International Conference on Robotics and Automation (ICRA), Paris, France, 31 May–31 August 2020; pp. 3770–3776. [ Google Scholar ]
• Qizilbash, A.A.H.; Henkel, C.; Mostaghim, S. Ant Colony Optimization based Multi-Robot Planner for Combined Task Allocation and Path Finding. In Proceedings of the 2020 17th International Conference on Ubiquitous Robots (UR), Kyoto, Japan, 22–26 June 2020; pp. 487–493. [ Google Scholar ]
• Han, Q. An Application of Improved PSO Algorithm in Cooperative Search Task Allocation. In Proceedings of the 2021 IEEE International Conference on Power Electronics, Computer Applications (ICPECA), Shenyang, China, 22–24 January 2021; pp. 580–583. [ Google Scholar ]
• Zhang, J.; Ren, J.; Cui, Y.; Fu, D.; Cong, J. Multi-USV Task Planning Method Based on Improved Deep Reinforcement Learning. IEEE Internet Things J. 2024 , 11 , 18549–18567. [ Google Scholar ] [ CrossRef ]
• Yi, X.; Zhu, A.; Yang, S.X.; Luo, C. A Bio-Inspired Approach to Task Assignment of Swarm Robots in 3-D Dynamic Environments. IEEE Trans. Cybern. 2017 , 47 , 974–983. [ Google Scholar ] [ CrossRef ]
• Sun, W.; Zhang, F.; Xue, M.; Hu, W.; Li, L. An SOM-based algorithm with locking mechanism for task assignment. In Proceedings of the 2017 IEEE International Conference on Cybernetics and Intelligent Systems (CIS) and IEEE Conference on Robotics, Automation and Mechatronics (RAM), Ningbo, China, 19–21 November 2017; pp. 36–41. [ Google Scholar ]
• Tang, H.; Lin, A.; Sun, W.; Shi, S. An Improved SOM-Based Method for Multi-Robot Task Assignment and Cooperative Search in Unknown Dynamic Environments. Energies 2020 , 13 , 3296. [ Google Scholar ] [ CrossRef ]
• Luo, Q.; Wang, H.; Li, N.; Zheng, W. Multi-unmanned surface vehicle model-free sliding mode predictive adaptive formation control and obstacle avoidance in complex marine environment via model-free extended state observer. Ocean Eng. 2024 , 293 , 116773. [ Google Scholar ] [ CrossRef ]
• Loria, A.; Dasdemir, J.; Jarquin, N.A. Leader–Follower Formation and Tracking Control of Mobile Robots Along Straight Paths. IEEE Trans. Control Syst. Technol. 2016 , 24 , 727–732. [ Google Scholar ] [ CrossRef ]
• Wang, W.; Tong, S.; Wang, D. Adaptive Fuzzy Containment Control of Nonlinear Systems With Unmeasurable States. IEEE Trans. Cybern. 2019 , 49 , 961–973. [ Google Scholar ] [ CrossRef ]
• Dong, X.; Hu, G. Time-Varying Formation Tracking for Linear Multiagent Systems with Multiple Leaders. IEEE Trans. Autom. Control 2017 , 62 , 3658–3664. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Wang, D.; Wang, W.; Liu, L. Containment control of networked autonomous underwater vehicles: A predictor-based neural DSC design. ISA Trans. 2015 , 59 , 160–171. [ Google Scholar ] [ CrossRef ]
• Liu, L.; Wang, D.; Peng, Z. Coordinated path following of multiple underacutated marine surface vehicles along one curve. ISA Trans. 2016 , 64 , 258–268. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Asadi, M.M.; Ajorlou, A.; Aghdam, A.G. Distributed control of a network of single integrators with limited angular fields of view. Automatica 2016 , 63 , 187–197. [ Google Scholar ] [ CrossRef ]
• Liang, X.; Qu, X.; Wang, N.; Li, Y.; Zhang, R. A Novel Distributed and Self-Organized Swarm Control Framework for Underactuated Unmanned Marine Vehicles. IEEE Access 2019 , 7 , 112703–112712. [ Google Scholar ] [ CrossRef ]
• Palacios, L.; Ceriotti, M.; Radice, G. Close proximity formation flying via linear quadratic tracking controller and artificial potential function. Adv. Space Res. 2015 , 56 , 2167–2176. [ Google Scholar ] [ CrossRef ]
• Liu, Y.-C.; Chopra, N. Control of semi-autonomous teleoperation system with time delays. Automatica 2013 , 49 , 1553–1565. [ Google Scholar ] [ CrossRef ]
• Oland, E.; Kristiansen, R. Collision and terrain avoidance for UAVs using the potential field method. In Proceedings of the 2013 IEEE Aerospace Conference, Big Sky, MT, USA, 2–9 March 2013; pp. 1–7. [ Google Scholar ]
• He, W.; Yin, Z.; Sun, C. Adaptive Neural Network Control of a Marine Vessel with Constraints Using the Asymmetric Barrier Lyapunov Function. IEEE Trans. Cybern. 2017 , 47 , 1641–1651. [ Google Scholar ] [ CrossRef ]
• Liang, X.; Qu, X.; Wan, L.; Ma, Q. Three-Dimensional Path Following of an Underactuated AUV Based on Fuzzy Backstepping Sliding Mode Control. Int. J. Fuzzy Syst. 2018 , 20 , 640–649. [ Google Scholar ] [ CrossRef ]
• Mu, D.; Wang, G.; Fan, Y.; Bai, Y.; Zhao, Y. Fuzzy-Based Optimal Adaptive Line-of-Sight Path Following for Underactuated Unmanned Surface Vehicle with Uncertainties and Time-Varying Disturbances. Math. Probl. Eng. 2018 , 2018 , 7512606. [ Google Scholar ] [ CrossRef ]
• Wang, N.; Er, M.J. Self-Constructing Adaptive Robust Fuzzy Neural Tracking Control of Surface Vehicles with Uncertainties and Unknown Disturbances. IEEE Trans. Control Syst. Technol. 2015 , 23 , 991–1002. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Wang, J.; Wang, D. Distributed Maneuvering of Autonomous Surface Vehicles Based on Neurodynamic Optimization and Fuzzy Approximation. IEEE Trans. Control Syst. Technol. 2018 , 26 , 1083–1090. [ Google Scholar ] [ CrossRef ]
• Cheng, Y.; Zhang, W. Concise deep reinforcement learning obstacle avoidance for underactuated unmanned marine vessels. Neurocomputing 2018 , 272 , 63–73. [ Google Scholar ] [ CrossRef ]
• Sui, Z.; Pu, Z.; Yi, J.; Wu, S. Formation Control with Collision Avoidance Through Deep Reinforcement Learning Using Model-Guided Demonstration. IEEE Trans. Neural Netw. Learn. Syst. 2021 , 32 , 2358–2372. [ Google Scholar ] [ CrossRef ] [ PubMed ]
• Wang, S.; Ma, F.; Yan, X.; Wu, P.; Liu, Y. Adaptive and extendable control of unmanned surface vehicle formations using distributed deep reinforcement learning. Appl. Ocean Res. 2021 , 110 , 102590. [ Google Scholar ] [ CrossRef ]
• Xie, J.; Zhou, R.; Liu, Y.; Luo, J.; Xie, S.; Peng, Y.; Pu, H. Reinforcement-Learning-Based Asynchronous Formation Control Scheme for Multiple Unmanned Surface Vehicles. Appl. Sci. 2021 , 11 , 546. [ Google Scholar ] [ CrossRef ]
• Chen, H.; Yan, H.; Wang, Y.; Xie, S.; Zhang, D. Reinforcement learning-based close formation control for underactuated surface vehicle with prescribed performance and time-varying state constraints. Ocean Eng. 2022 , 256 , 111361. [ Google Scholar ] [ CrossRef ]
• Wu, C.; Zhao, M.; Min, C.; Wang, Y.; Luo, J. Adaptive neural formation control of autonomous underactuated surface vessels based on disturbance observer with leader–follower strategy. Trans. Inst. Meas. Control 2021 , 43 , 2972–2982. [ Google Scholar ] [ CrossRef ]
• Ghommam, J.; Saad, M.; Mnif, F.; Zhu, Q.M. Guaranteed Performance Design for Formation Tracking and Collision Avoidance of Multiple USVs With Disturbances and Unmodeled Dynamics. IEEE Syst. J. 2021 , 15 , 4346–4357. [ Google Scholar ] [ CrossRef ]
• He, S.; Wang, M.; Dai, S.L.; Luo, F. Leader–Follower Formation Control of USVs With Prescribed Performance and Collision Avoidance. IEEE Trans. Ind. Inform. 2019 , 15 , 572–581. [ Google Scholar ] [ CrossRef ]
• Chen, G.; Yao, D.; Zhou, Q.; Li, H.; Lu, R. Distributed Event-Triggered Formation Control of USVs with Prescribed Performance. J. Syst. Sci. Complex. 2022 , 35 , 820–838. [ Google Scholar ] [ CrossRef ]
• Dai, S.L.; He, S.; Lin, H.; Wang, C. Platoon Formation Control with Prescribed Performance Guarantees for USVs. IEEE Trans. Ind. Electron. 2018 , 65 , 4237–4246. [ Google Scholar ] [ CrossRef ]
• Huang, C.; Zhang, X.; Zhang, G. Improved decentralized finite-time formation control of underactuated USVs via a novel disturbance observer. Ocean Eng. 2019 , 174 , 117–124. [ Google Scholar ] [ CrossRef ]
• Zhang, G.; Liu, S.; Huang, J.; Zhang, W. Dynamic Event-Triggered Path-Following Control of Underactuated Surface Vehicle with the Experiment Verification. IEEE Trans. Veh. Technol. 2022 , 71 , 10415–10425. [ Google Scholar ] [ CrossRef ]
• Xiang, X.; Yu, C.; Lapierre, L.; Zhang, J.; Zhang, Q. Survey on Fuzzy-Logic-Based Guidance and Control of Marine Surface Vehicles and Underwater Vehicles. Int. J. Fuzzy Syst. 2018 , 20 , 572–586. [ Google Scholar ] [ CrossRef ]
• Zhou, W.; Wang, Y.; Ahn, C.K.; Cheng, J.; Chen, C. Adaptive Fuzzy Backstepping-Based Formation Control of Unmanned Surface Vehicles with Unknown Model Nonlinearity and Actuator Saturation. IEEE Trans. Veh. Technol. 2020 , 69 , 14749–14764. [ Google Scholar ] [ CrossRef ]
• Wu, T.; Xue, K.; Wang, P. Leader-follower formation control of USVs using APF-based adaptive fuzzy logic nonsingular terminal sliding mode control method. J. Mech. Sci. Technol. 2022 , 36 , 2007–2018. [ Google Scholar ] [ CrossRef ]
• Liu, L.; Wang, D.; Peng, Z.; Chen, C.L.P.; Li, T. Bounded Neural Network Control for Target Tracking of Underactuated Autonomous Surface Vehicles in the Presence of Uncertain Target Dynamics. IEEE Trans. Neural Netw. Learn. Syst. 2019 , 30 , 1241–1249. [ Google Scholar ] [ CrossRef ]
• Zhang, S.; Xiang, X.; Yang, S. Cooperative control system of multi-ASV: Design and implementation. In Proceedings of the 2018 IEEE 8th International Conference on Underwater System Technology: Theory and Applications (USYS), Wuhan, China, 1–3 December 2018; pp. 1–6. [ Google Scholar ]
• Liu, K.; Fridman, E.; Hetel, L. Networked Control Systems in the Presence of Scheduling Protocols and Communication Delays. SIAM J. Control Optim. 2015 , 53 , 1768–1788. [ Google Scholar ] [ CrossRef ]
• Gao, J.; Zhang, G.; Wu, P.; Zhao, X.; Wang, T.; Yan, W. Model Predictive Visual Servoing of Fully-Actuated Underwater Vehicles with a Sliding Mode Disturbance Observer. IEEE Access 2019 , 7 , 25516–25526. [ Google Scholar ] [ CrossRef ]
• Wang, N.; Sun, Z.; Yin, J.; Zou, Z.; Su, S.-F. Fuzzy unknown observer-based robust adaptive path following control of underactuated surface vehicles subject to multiple unknowns. Ocean Eng. 2019 , 176 , 57–64. [ Google Scholar ] [ CrossRef ]
• Zereik, E.; Bibuli, M.; Mišković, N.; Ridao, P.; Pascoal, A. Challenges and future trends in marine robotics. Annu. Rev. Control 2018 , 46 , 350–368. [ Google Scholar ] [ CrossRef ]
• Jiang, Y.; Peng, Z.; Wang, D.; Chen, C.L.P. Line-of-Sight Target Enclosing of an Underactuated Autonomous Surface Vehicle with Experiment Results. IEEE Trans. Ind. Inform. 2020 , 16 , 832–841. [ Google Scholar ] [ CrossRef ]
• Fan, J.; Li, Y.; Liao, Y.; Ma, T.; Ge, Y.; Wang, Z. A Formation Reconfiguration Method for Multiple Unmanned Surface Vehicles Executing Target Interception Missions. Appl. Ocean Res. 2020 , 104 , 102359. [ Google Scholar ] [ CrossRef ]
• Miao, R.; Wang, L.; Pang, S. Coordination of distributed unmanned surface vehicles via model-based reinforcement learning methods. Appl. Ocean Res. 2022 , 122 , 103106. [ Google Scholar ] [ CrossRef ]
• Wang, C.; Zhang, H.; Yu, Y. USV trajectory tracking control system based on ADRC. In Proceedings of the 2017 Chinese Automation Congress (CAC), Jinan, China, 20–22 October 2017; pp. 7534–7538. [ Google Scholar ]
• Liu, L.; Wang, D.; Peng, Z. Path following of marine surface vehicles with dynamical uncertainty and time-varying ocean disturbances. Neurocomputing 2016 , 173 , 799–808. [ Google Scholar ] [ CrossRef ]
• Wang, N.; Gao, Y.; Sun, Z.; Zheng, Z. Nussbaum-Based Adaptive Fuzzy Tracking Control of Unmanned Surface Vehicles with Fully Unknown Dynamics and Complex Input Nonlinearities. Int. J. Fuzzy Syst. 2018 , 20 , 259–268. [ Google Scholar ] [ CrossRef ]
• Peng, Z.; Wang, D.; Li, T.; Wu, Z. Leaderless and leader-follower cooperative control of multiple marine surface vehicles with unknown dynamics. Nonlinear Dyn. 2013 , 74 , 95–106. [ Google Scholar ] [ CrossRef ]
• Qu, Y.; Xiao, B.; Fu, Z.; Yuan, D. Trajectory exponential tracking control of unmanned surface ships with external disturbance and system uncertainties. ISA Trans. 2018 , 78 , 47–55. [ Google Scholar ] [ CrossRef ]
• Li, R.; Huang, J.; Pan, X.; Hu, Q.; Huang, Z. Path following of underactuated surface ships based on model predictive control with neural network. Int. J. Adv. Robot. Syst. 2020 , 17 , 1729881420945956. [ Google Scholar ] [ CrossRef ]
• Weng, Y.; Wang, N.; Qin, H.; Karimi, H.R.; Qi, W. Data-Driven Adaptive Tracking Control of Unknown Autonomous Marine Vehicles. IEEE Access 2018 , 6 , 55723–55730. [ Google Scholar ] [ CrossRef ]
• Zheng, Z.; Sun, L. Path following control for marine surface vessel with uncertainties and input saturation. Neurocomputing 2016 , 177 , 158–167. [ Google Scholar ] [ CrossRef ]
• Liao, Y.; Jiang, Q.; Du, T.; Jiang, W. Redefined Output Model-Free Adaptive Control Method and Unmanned Surface Vehicle Heading Control. IEEE J. Ocean. Eng. 2020 , 45 , 714–723. [ Google Scholar ] [ CrossRef ]
• Xu, H.; Hassani, V.; Guedes Soares, C. Uncertainty analysis of the hydrodynamic coefficients estimation of a nonlinear manoeuvring model based on planar motion mechanism tests. Ocean Eng. 2019 , 173 , 450–459. [ Google Scholar ] [ CrossRef ]
• Shen, H.; Wen, G.; Lv, Y.; Zhou, J.; Wang, L. USV Parameter Estimation: Adaptive Unscented Kalman Filter-Based Approach. IEEE Trans. Ind. Inform. 2023 , 19 , 7751–7761. [ Google Scholar ] [ CrossRef ]

Click here to enlarge figure

Share and Cite

Xue, J.; Song, Y.; Hu, H. Formation Control of a Multi-Unmanned Surface Vessel System: A Bibliometric Analysis. J. Mar. Sci. Eng. 2024 , 12 , 1484. https://doi.org/10.3390/jmse12091484

Xue J, Song Y, Hu H. Formation Control of a Multi-Unmanned Surface Vessel System: A Bibliometric Analysis. Journal of Marine Science and Engineering . 2024; 12(9):1484. https://doi.org/10.3390/jmse12091484

Xue, Jie, Yuanming Song, and Hao Hu. 2024. "Formation Control of a Multi-Unmanned Surface Vessel System: A Bibliometric Analysis" Journal of Marine Science and Engineering 12, no. 9: 1484. https://doi.org/10.3390/jmse12091484

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals

Multi-agent platform to support trading decisions in the FOREX market

• Open access
• Published: 30 August 2024

Cite this article

You have full access to this open access article

• Marcin Hernes   ORCID: orcid.org/0000-0002-3832-8154 1 ,
• Jerzy Korczak 2 ,
• Dariusz Krol 3 ,
• Maciej Pondel 1 &
• Jörg Becker 4  

Trading decisions often encounter risk and uncertainty complexities, significantly influencing their overall performance. Recognizing the intricacies of this challenge, computational models within information systems have become essential to support and augment trading decisions. The paper introduces the concepts of trading software agents, investment strategies, and evaluation functions that automate the selection of the most suitable strategy in near real-time, offering the potential to enhance trading effectiveness. This approach holds the promise of significantly increasing the effectiveness of investments. The research also seeks to discern how changing market conditions influence the performance of these strategies, emphasizing that no single agent or strategy universally outperforms the rest. In summary, the overarching objective of this research is to contribute to the realm of financial decision-making by introducing a pragmatic platform and strategies tailored for traders, investors, and market participants in the FOREX market. Ultimately, this endeavor aims to empower people with more informed and productive trading decisions. The contributions of this work extend beyond the theoretical realm, demonstrating a commitment to address the practical challenges faced by traders and investors in real-time decision-making within the financial markets. This multidimensional approach to financial decision support promises to enhance investment effectiveness and contribute to the broader field of algorithmic trading.

Similar content being viewed by others

On Evolving Multi-agent FX Traders

AURORA: an autonomous agent-oriented hybrid trading service

Enhancing Trading Strategies: A Multi-indicator Analysis for Profitable Algorithmic Trading

Explore related subjects.

• Artificial Intelligence

Avoid common mistakes on your manuscript.

1 Introduction

A combination of statistical analysis, financial mathematics, econometrics, and, increasingly, artificial intelligence often informs trading decisions. These methods are frequently integrated into multi-agent systems to enhance trading activities in the foreign exchange market (FOREX) [ 1 ]. These systems strongly emphasize high-frequency trading (HFT), short-term position openings/closings, and sophisticated algorithms that leverage robust indicators and modern technology. The goal is to generate profits by capitalizing on minimal price fluctuations, characterized by high-frequency occurrences, where profits often arise from market liquidity imbalances or short-term pricing inefficiencies.

In general, the trading platforms must offer real-time guidance on trading positions, such as when to open/close positions, whether to go long or short or when to step away from investments. These guidelines form specific trading strategies, defined by their verifiability, quantifiability, consistency, and objectivity [ 2 ].

A trading strategy should outline the assets, entry / exit points, and money management rules, drawing from fundamental, technical, or behavioral analysis. These strategies are validated through backtesting (historical data) or forward testing (simulated trading environments compared to real-world results). Online trading adds further challenges [ 3 ], requiring the real-time use of one or multiple algorithms, often implemented as software agents. Currently, most trading systems are based on single or numerous algorithms without employing agents [ 4 , 5 , 6 , 7 , 8 , 9 , 10 ]. They are also based on the single agent architecture [ 11 ]. This paper introduces A-Trader, a multi-agent platform designed to support financial decision-making within the FOREX market. As the A-Trader platform is presented, several critical issues in designing advisory systems for stock markets will be addressed. These challenges encompass:

Integration of Diverse Decision Sources: Harmonizing many decision sources, offering insight into effectively integrating varied inputs for informed decision-making.

Selection of Recommendation Methods and Algorithms: Exploring the crucial task of selecting the most suitable recommendation methods and algorithms.

Cooperation and Control of Advisory Algorithms: The importance of seamless collaboration and control of advisory algorithms is emphasized, providing valuable insights into optimizing algorithmic efficiency.

Composition of the Global Investment Strategy Evaluation Criterion: We underscore a holistic approach to performance assessment by examining the need for a comprehensive evaluation criterion for global investment strategies.

System Openness and Interoperability: The authors discuss the importance of system flexibility and adaptability, offering an in-depth understanding of the prerequisites for system openness.

The solutions implemented in the A-Trader platform will exemplify the issues mentioned above. A-Trader is a dynamic multi-agent experimental platform for constructing, simulating, and assessing investment strategies, catering to various investor types. Technically, A-Trader is integrated with an online data system, MetaTrader, which provides raw and preprocessed data and buy-sell decisions generated by agents using various methods. The platform develops investment strategies and continuously evaluates them based on the open/close and short/long positions determined by the most highly rated agents. The significant advantage of A-Trader over other trading platforms lies in its use of a multicriteria function to evaluate the strategy unlike platforms that rely solely on return-based metrics, A-Trader calculates a return rate based on risk-based measures, including factors like the number of transactions, gross profit, gross loss, profitable trades, consecutive profitable transactions, non-profitable successive transactions, Sharpe ratio, average volatility coefficient, and average return per transaction [ 12 , 13 ].

In this paper, it will be demonstrated that:

The use of advanced technologies and a system architecture offers better performance and greater openness than existing solutions.

Provides a flexible and agile methodology for the development of investment strategies.

It ensures more realistic trading performance analysis based not only on return-based metrics but also on risk assessment and endogenous benchmarks.

The approach allows for the creation of strategies with superior performance compared to other methods.

The multi-agent approach enables the simulation of trader behavior, which can be used to enhance FOREX decision-making processes.

The paper is structured as follows. The first part of the paper introduces A-Trader’s architecture and functionalities. The second part delves into the specifications of various trading agents. Three categories of trading agents are examined: agents based on technical analysis, agents based on macroeconomic and fundamental analysis, and behavior-based agents. Subsequently, it outlines trading strategy-building approaches using the set of available agents, and concludes with an analysis of the results from research experiments evaluating the performance of selected trading strategies on FOREX.

2 Background

The design and implementation of multi-agent systems in stock trading has been a focal point for numerous projects and research reports.

2.1 Multi-agent systems for financial decision support

This paper [ 14 ] proposes a modular multi-agent reinforcement learning-based system for financial portfolio management (MSPM) to address the challenges of scalability and reusability in adapting to ever-changing markets. Using evolving agent modules (EAMs) for generating information and Strategic Agent Modules (SAMs) for portfolio optimization, the system ensures improved adaptability and performance, evidenced by significant outperformance in US stock market data. The multi-agent deep reinforcement learning framework proposed in [ 15 ] leverages the collective intelligence of expert traders, each focused on different timeframes, to improve trading outcomes. It employs a hierarchical structure in which knowledge flows from agents trading on higher time frames to those on lower time frames, improving robustness against noise in financial data. Other examples of multi-agent architectures based on the deep reinforcement learning framework are shown in papers [ 16 ] and [ 17 ].

A proposal for a framework for evolutionary multi-agent trading for FOREX was introduced in [ 18 ]. In this paper, the authors focused on currency trading and included the impact of FX trading spread. They used technical indicators to provide temporary features from which a decision tree defined the training strategy. Tree representation classifiers were built with Genetic Programming (an evolutionary technique). The authors proposed a general FOREX Genetic Programming Framework (FXGP), and the proposed simplified framework (sFXGP) has been deployed to construct multiple agents operating concurrently [ 19 ]. The works [ 20 , 21 ] present an approach for financial market prediction, where agents examine the similarities between the ask and bid asset histories to predict quotes in real time. The paper [ 22 ] shows development of ForexMA, a multi-agent system that enhances decision-making in Forex trading by integrating both qualitative and quantitative information. The architecture includes three agents, namely, the Facts Analyzing Agent, the Decision Agent, and the Performance Analyzing Agent. The authors demonstrated that ForexMA outperforms human expert traders by delivering high-frequency, rapid solutions in a matter of seconds. This system was tested and proven to generate more accurate predictions than those made by human experts, who typically operate on lower frequency timeframes and require several hours to analyze the information.

2.2 Advanced methods for financial decision supporting

This section analyses the methods developed not as agent-based approaches but can be transformed into agent structures in multi-agent systems.

The works [ 23 , 24 ] present the use of neuro-fuzzy computing and neural networks for making quotation predictions based on analysis of a financial time series’s geometrical patterns. Another paper proposing a behavioral approach for trading decisions is [ 25 ]. Some authors present strategies based on trading bots [ 26 ] or deep belief networks (DBN) [ 27 ] to build investment decisions based on the S&P500. Deep learning techniques, in turn, are presented in [ 28 ]. The deep learning approach is based on such methods, as recurrent neural networks, including Long Short-Term Memory [60], spiking neural networks [ 29 , 30 , 31 ]. Machine learning (ML) techniques significantly impact on the automatic identification of trading agents to identify profitable strategies to trade in the stock or currency market. Financial predictions incorporating ML approaches construct training, test, and off-sample data sets as a collection of instances using commonly used technical indicators. An example of ML models applied to trading scenarios in the FOREX market was discussed in [ 32 ]. The authors wanted to verify whether, using these models, it is possible to obtain consistently profitable returns. The authors proved that while getting good returns using simple classifiers is possible, each model needed a specific setup, including variables such as the retraining period, the size of the retraining set, and the number and type of attributes selected to construct the model. The complexities of the market require a combination of parameters that, for the same instrument, could change under different market conditions and seasons. The models needed to learn new patterns to cope with the dynamics of the market, but at the same time, to avoid noisy ways that might not be related to the current market situation could only be based on training comprised of current values of the time series using a sliding window approach.

An ensemble approach was proposed in [ 33 ] where the authors classify the FOREX market, and behavioral analyzes are considered by a certain amount, 2) downtrends when FX rates decrease by a certain amount, and 3) sideways trends. They extract features from these trends using multi-scale features. Multiple classifiers are trained using these features. Bayesian voting was used to create an ensemble of these classifiers, which can recognize trends in the market. The experimental results showed that the proposed system could accurately identify up and down trends in the FX rate signal.

Mayo states that a significant amount of intraday market data is noise or redundant, and if it is eliminated, then predictive models built using the remaining intraday data will be more accurate. He proposed an algorithm known as Evolutionary Data Selection (EDS), which uses a model building algorithm in conjunction with the available training data to find an optimal subset of those data [ 34 ].

Until now, articles have discussed the competition between multi-agent trading systems and their performance in trading scenarios [ 50 ]. Some of them explore advances in artificial economics, including agent-based models, and their applications in finance and game theory [ 51 ]. Focusing on the evolution of multi-agent foreign exchange (FX) traders, Longinov analyzes their performance in FX markets [ 18 ]. Currently, there are many platforms for HFT decision support in FOREX, such as FinEXo, Trade360, AvaTRADE, EXsignals, and Trade Chimp.

2.3 Assessment of existing approaches

The presented theoretical and practical approaches and solutions are often insufficient for HFT decision support. They are characterized by low performance (insufficient to support HFT) and costly maintenance. Moreover, the problem with openness and integration of the technologies appears in most cases.

The existing platforms are mainly based on technical analysis. Fundamental analysis and behavioral analysis are considered to a low degree. The disadvantage of existing approaches is also a performance measurement process. Mainly return-based measures are only taken into consideration, and it causes to limitation of personalization of the strategies evaluation byways(for example, a specific group of users can take into consideration mainly the rate of return-based measures, and other groups of users want to take into consideration combining-based measures). Therefore, investors often also need other classes of measures, for example, risk-based measures, to properly manage risk. The existing platforms are also not fully open-accessible, and users can develop their strategies using the tools proposed by the given platform. It is very difficult to integrate strategies developed by a user in other software environments with the given trading platform. Therefore, the main research problem undertaken in this paper is to develop an approach that overcomes the presented disadvantages of existing approaches. For this purpose, we developed the conception and prototype of a multi-agent platform in our research.

3 Architecture and functionalities of A-trader

The primary strengths inherent in multi-agent systems, including A-Trader, lie in their openness to integrate novel trading algorithms and specific functionalities that enable model-building communication among various agents. These systems operate on the principles of collective intelligence, allowing for tailored solutions using diverse market monitoring methods. Multi-agent technology facilitates the customization of solutions through agents that evaluate existing methods and preprocess learning datasets. These agents have learning capabilities, evolving their knowledge about financial market behavior. Overcoming computational challenges are achieved by leveraging a service-oriented architecture and cloud computing (SOAP). The SOAP communication protocol, as implemented in A-Trader, greatly simplifies the integration of individual solutions due to its open and easily implementable nature Footnote 1 . Incorporating PUSH technology, a common feature in distributed systems, notably accelerates information propagation within the system, as discussed in further detail in [ 35 ]. The system retrieves real-time data from the currency market using MetaTrader or JForex software. A-Trader analyzes quotation data using many criteria, ensuring near real-time processing and the capability to handle diverse data sources. For a more in-depth understanding of A-Trader’s architecture, system elements, and agent details, refer to [ 13 , 36 , 37 ].

In general terms, A-Trader is composed of agents capable of generating independent decisions. These decisions can be characterized by model building by consistency or contradiction, e.g., the two independent agents can simultaneously generate open and closed positions. Figure  1 presents an overview of the architecture and functional concept of A-Trader.

A-Trader system architecture

The main goal of the Supervisor Agent (SA) is to generate profitable trading advice to achieve a specific rate of return and reduce investment risk. This agent performs based on Basic and Intelligent agents’ decisions. It provides different trading strategies and final open/close long/short positions to the trader or automatically to the market. The Supervisor also resolves Computing Agent knowledge conflicts within the Cloud and evaluates their performance. Based on collected knowledge, this agent determines which decisions are considered in a given strategy and which are ignored.

The Notification Agent (NA) receives the data (quotations), distributes messages (signals) to various agents, and controls the system operation running in a multi-threaded manner. Information about the message flow (which agent sends signals to which agent) is read during the NA initialization from the Routing Table.

Figure  2 shows an example of the data flow inside the NA. This agent “listens” at the given port, and if information from Agent A5 is received, then NA searches, in the Routing Table, the agents who listen to messages (signals) from Agent A5. In the considered example, these are Agents A7 and A9. Next, the NA agent searches for threads being sent (Sending Threads Table) to Agents A7 and A9 and sends them through.

Data flow inside the Notification Agent

The Cloud of Computing Agents (CCA) consists of the Basic Agents Cloud (BAC) and Intelligent Agents Cloud (IAC). BAC consists of agents that preprocess the data and calculate the fundamental technical analysis indicators. IAC consists of agents with a knowledge base. They can perform the learning process and can change their internal state and parameters. This group of agents uses methods based on artificial intelligence (neural networks, rule-based systems, genetic algorithms, cognitive technologies, etc.), agents observing market behavior and agents analyzing text messages. User-defined Intelligent Agents Cloud (UAC) consists of agents created by external users. Integration of User-defined Agents within the system without installing the agent on the servers is possible in A-Trader. The result of the Basic Agents and the Intelligent Agents activity is a decision that the NA transfers to the Supervisor Agent.

The Market Communication Agents (MCA) communicate between A-Trader and the external environment. MCA provides the actual values of quotations, and they are responsible for performing open/close long/short position orders.

A visualization agent (VA) visualizes quotations, decisions, and long/short positions in the form of charts.

The layer of Cloud Computing Agents is the system’s core that analyzes signals contained in notifications and delivers decision recommendations to the Supervisor Agent. The Supervisor Agent then generates the final decision, as previously stated. Selected agents (especially belonging to CCA) running on A-Trader architecture are described in the next section.

Analyzing the computational complexity of a-Trader, it should be noted that it depends on the computational complexity of the algorithms of the individual agents. However, the architecture of the system makes it possible to determine decisions within 5 to 20 milliseconds of receiving the last quotation as an input signal (server parameters: Intel Core i7-9700K 8 cores, RAM 16 GB, NVIDIA GeForce RTX 2060 16GB, SSD M.2 480 GB, HDD SATA 7200 2000 GB).

4 Agent descriptions

A software agent is an intelligent program that not only executes based on acquired data but also takes specific actions to achieve a specified goal (for example, making satisfactory decisions in the FOREX market). A-Trader contains various types of agent, as mentioned in the previous chapter: Market Communication Agents, Notification Agents, Visualization Agents, Supervisor Agents, Historical Agents, and agents belonging to the cloud of computing agents (Basic Agents, Intelligent Agents, and User Agents), and currently, approximately 1600 agents are implemented on the platform. In cloud of computing agents These there are 800 basic agents (BAC) processing data agents (these agents calculate mainly technical analysis indicators related to FOREX market quotations), 500 intelligent agents (IAC), running in different aggregates and generating buy-sell decisions (about 250 agents based on three-valued logic, 250 agents based on fuzzy logic) and 300 agents generating open/close long/short positions (thus providing the strategies).

An experimental platform was designed to easily integrate new agents (user agents) and allow the reuse of existing agent decisions in new strategies. We adopted three conventions for generating agent responses / signals: three-value logic, fuzzy logic, and our signals. Three-value logic is a manner for the representation of agents’ knowledge to provide buy / sell decision signals, generated as the agent’s output signal, where the value 1 denotes a buy decision, the value -1 denotes a sell decision, and the value 0 denotes don’t care . For a trading decision, fuzzy logic agents are more appropriate. The confidence range for decisions on A-Trader is \([-1\dots 1]\) , where ’-1’ denotes a strong sell decision, ’0’ denotes a strong leave unchanged decision and ’1’ denotes a strong buy decision. The signal for open/close positions can then be generated based on a given decision’s confidence level. For example, a short position is opened when a confidence level is greater than -0.8, whereas a long position is opened when a confidence level is more significant than 0.6. As a result, open/closed positions can achieve more profitable results than positions generated based on three-value logic. It should be stressed that the level of confidence for open/close positions is very important, and it can be determined by considering trader experience or automatically determined by the Supervisor using, for example, a genetic algorithm. Specific signals are generated by agents which do not have simple/linear interpretation, for example, signals from agents with unsupervised learning.

Currently, A-Trader consists of three groups of buy/sell decision agents.

4.1 Agents based on technical analysis

Agents based on technical analysis use three-valued logic or fuzzy logic. Technical indicators have interpretations such as the market is oversold, the power of buyers is exhausted, etc. where assembling some of these may give satisfactory results. The shorter the investment horizon, the greater the effectiveness of technical analysis. To illustrate how an agent works, let us present an example of a fuzzy logic agent called FuzzyTrendLinearRegression . This agent makes decisions in the following manner. A given number of M quotations is approximated by the equation: \(y = ax + b\) (straight line). The inclination of this line depends on the value of the coefficient “a” or the tangent value of the inclination angle using linear regression.

The FuzzyTrendLinearRegression agent specification.

The agent generates a buy signal when the coefficient value of “a” changes from positive to negative, and it generates a sell signal when the coefficient “a” changes from negative to positive. The transition of the agent’s decision is performed using the hysteresis level, defined by the coefficient value \(\delta \) .

4.2 Agents based on macro-economic and fundamental analysis

A-Trader also consists of agents based on fundamental analysis and behavioral data. The fundamental analysis in FOREX is related to the economic, social, and political forces driving demand and supply on the currency market. The level of the supply and demand balance is affected by two main factors:

Interest rates can strengthen or weaken a particular currency where a high level of interest rates (as compared to those in other currencies) can increase the level of foreign investment in a currency, which in turn, leads to a strengthening of the currency.

The international trade balance deficit (higher value of imports than the value of exports) can usually adversely affect a currency. In this case, the currency is transferred out of a country to buy foreign products, which can lead to a devaluation of the currency.

Other factors, such as central bank interventions (e.g., by increasing / reducing foreign exchange reserves) strengthen / reduce demand for a specific currency. Fundamental analysis is based on an examination of asset markets, macroeconomic indicators, and political considerations of the country to evaluate the development of the exchange rate of a particular currency. Asset markets include stock exchanges, bond markets, and real estate. Macroeconomic indicators are measured by Gross Domestic Product, Money Supply (M1, M5, D1, W1, etc.), unemployment, inflation, foreign exchange reserves, interest rates, and productivity. Political considerations can influence the level of certainty of stability and the level of confidence in a nation’s government. The fundamental analysis agents also consider indicators such as the Consumer Price Index (CPI), Durable Goods Orders, Producer Price Index (PPI), Purchasing Managers Index (PMI) and retail sales.

However, often online fundamental analysis only sometimes provides market entry and exit points in FOREX as a lot of information emerges at regular intervals. Still, only a part of this information is relevant. Therefore, there are only a few agents based on macro-economic and fundamental analysis are implemented, notably:

Interest rates - if interest rates are higher in one country than in its neighbors, the currency prices in this country will often strengthen because a higher interest rate attracts more foreign investors.

Gross Domestic Product (GDP) is the sum of all goods and services produced/provided by domestic or foreign companies in a given country. Based on GDP, the level of growth (or contraction) of a country’s economy can be measured. This indicator has the broadest scope for the change in economic output and production in a given country. The Gross National Product (GNP), in turn, is related to the nationality of capital.

Purchasing Manager’s Index (PMI) includes data related to new orders, supplier delivery times, production, backlogs, prices, inventories, employment, import and export orders. It is characterized by high correlation with Monetary Policy Decisions and is a valuable tool to track the health of a country’s manufacturing sector.

S&P 500 is treated as a leading indicator of US equities and is meant to reflect the return/risk characteristics of the large cap universe, this index includes 500 stocks chosen on the basis of market size, liquidity industry grouping, and other factors.

FTSE 100 is a London Stock Exchange indicator and includes 100 companies characterized by the highest market capitalization on this Exchange.

WIG–is a Warsaw Stock Exchange index that includes securities listed on the main market.

To illustrate one of the agents based on macro-economic analysis, there is an agent called FuzzyNeuralNetIndices . The agent computes by applying Multilayer Perceptron to the trading decisions on the S&P500 and WIG indices.

The FuzzyNeuralNetIndices agent specification.

This agent is based on the interpretation of the money flow. If WIG20 is rising and the S&P 500 is falling, it can be predicted that investors can exchange their S&P shares for USD, then they can exchange USD for PLN to finally buy WIG shares. Therefore, if they buy PLN for USD, the value of PLN about USD should grow. Other fundamental analysis agents consider information about:

Gold prices ratio: when the price of gold goes down, then the USD often goes up (and vice versa); that means that prices of gold tend to have an inverse relationship to the price of USD and currency traders can take advantage of this relationship.

Oil price ratio: economies of oil-dependent countries grow (investors buy their currencies as a consequence) as oil prices drop.

4.3 Behavior-based agents

Many experts point out that the currency market is strongly correlated with the expectations of traders and their assessment of these expectations. There is a commonly observed relationship between stock prices and the behavior of traders, notably their perception of risk and benefit. Various prognoses, bulletins, and blogs strongly influence these expectations. An understanding of investor psychology can generate profit opportunities and thus can be extremely valuable for designing trading strategies. Many studies of behavioral models are used in FOREX trading, most based on psychology theories and applying data mining methods [ 38 ]. However, to validate these models on real financial markets, detailed information about traders, their experience and knowledge, and their psychological biases is needed.

The specification of the agent working on SENTIMENT index values.

Considering the limited sources of information on these subjects, in A-Trader only a behavioral time series has been provided and a few behavioral agents have been implemented [ 20 , 39 ]. The datasets are a broad range of day-by-day indicators (sentiments) provided by Polands MarketPsych Data or INI indicator. The indicators have been computed from millions of articles and posts in the news and on social media. In the experiments, behavioral indicators such as SENTIMENT, OPTIMISM, FEAR, as they relate to specific countries and their currencies (e.g. USD/PLN) are updated every day for countries and currencies and are input directly into A-Trader agents. For example, the SENTIMENT index indicates the 24-hour rolling average score of references in news and social networks to overall positive references, net of negative references. The OPTIMISM index is a bipolar emotional indicator in the range of -1 to 1. For interpretation purposes, gradual improvement of the SENTIMENT drives the continuation of the trend.

As mentioned above, agents can generate decisions that may be mutually consistent or completely contradictory. In A-Trader, the conflicts between agents are resolved by the Supervisor. This agent receives signals from decision-making agents and evaluates their performance. Through this evaluation, the Supervisor determines the agents for building investment strategies. In this way, the Supervisor can apply various strategies to generate open/close long/short position signals. The following section describes examples of these strategies.

5 Trading strategy construction

The strategies of A-Trader are built on the basis of the following assumptions:

Buy/sell decisions generated by a Cloud of Computing Agents form a base for strategy building. Every agent running in this Cloud sends its decision to the NA based on a unique decision method for each agent.

The Supervisor Agent builds investment strategies based on buy/sell decisions generated by Cloud Computing Agents (read from NA). These strategies generate the open/close long/short position signals.

Users–traders or bots (automatic traders), who invest in FOREX.

The strategies of A-Trader are based on more complex algorithms than algorithms based on technical analysis indicators [ 40 ] and, to illustrate the applied concept, four strategies are detailed: MyStrategy , Consensus , Evolution-based , Deep learning . These strategies have been chosen, because they were developed on the basis of the deep literature study and based on many experiments (these strategies are in the advanced phase of development in A-Trader, and the remaining strategies are in the preliminary phase of development).

5.1 MyStrategy

The strategy called MyStrategy is built of the basis of the following technical analysis, fundamental analysis, and behavior-based agents’ signals:

FuzzyRSI based on the Relative Strength Index indicator,

FuzzyROC based on the Rate of Change indicator,

FuzzyCCI based on the Commodity Channel Index indicator,

FuzzyMACD based on the Moving Average Convergence Divergence indicator,

FuzzyBollinger based on the Bollinger Bands indicator,

FuzzyWilliams based on the Williams %R indicator,

FuzzyNeuralNetIndices,

BehavioralAgent.

This strategy is run so that the open / close short / long position signal is generated when the average of fuzzy agent signals is higher / lower than a predefined threshold.

The strategy can be defined as follows:

The specification of MyStrategy.

5.2 Consensus strategy

The strategy Consensus , built on developing a consensus that determines the issues for financial decisions, is described in detail in [ 41 , 42 ]. The consensus agent, presented in detail in [ 36 ], develops a trading strategy based on a set of decisions generated by fuzzy logic agents.

The strategy can be specified as follows:

The specification of Consensus.

5.3 Evolution-based strategy

The strategy Evolution-based is developed based on work [ 52 ]. This strategy determines the best thresholds for open/close long/short positions based on decisions generated by technical analysis agents, fundamental analysis agents, and behavior-based agents. The Evolution-based strategy determines which agents should be considered when generating long/closed open/short position signals. It also determines the importance of decisions generated by a specific agent. The evolutionary algorithm indicates the space of agent decisions and weights their importance. The genotype in Fig.  3 consists of the weightings and thresholds for the opening / closing of the short / long position for each agent separately.

Genotype used in the Evolution-based strategy

In addition to weighting and thresholds, every advisory agent is characterized by ’compulsory’ parameters. These parameters mean that the agent’s signal value must be open, close, or ’don’t care’. The genotype also consists of values such as Profit Taking, Trailing Stop and Stop Loss for long and short positions. The result of this algorithm is a phenotype - a set of decision rules. For example, the open short position rules for the agent at time \(T_0\) can be specified as follows:

where: \(A_n T_0\) – value of Agent n signal in time \(T_0\) ,

\(w_so_n\) – weighting for Agent n short position opening,

\(Th_so\) – threshold for Agent n short position opening,

\(C_so_n\) – compulsory parameter for Agent n short position opening.

The conditions for the open/close short/long position are divided into two parts. The algorithm checks if a threshold is reached in the first part. The threshold is checked by multiplying the signals of each agent by the corresponding weightings, then all the results are to be summed up. The first part of the condition is met if the sum is higher than the opening short position threshold. The algorithm checks if all the mandatory rules are met in the second part of the condition. If a compulsory parameter of Agent 1 (OSO1) is equal to zero, the algorithm ’does not care’ what the value of Agent 1 is. If the parameter is equal to 1, the condition will be fulfilled only when the signal value of Agent 1 is positive. Similarly, in the case where the compulsory parameter is equal to -1, the algorithm expects a negative value of Agent 1. The compulsory parameters are checked for every advising agent. The strategy can be specified as follows.

The specification of Evolution-based.

5.4 Deep learning strategy

The  Deep learning strategy has been implemented on an open-source \(H_2O\) platform [ 24 ]. It is a distributed, scalable, and interactive in-memory data analysis and modeling solution. This platform consists of several data analysis models, including the Deep Learning Model, for Big Data exploration. In our approach, \(H_2O\) has been integrated with A-Trader.

Schema of Deep Learning strategy

The DeepLearning H \(_2O\) Agent is controlled by Supervisor and runs in two modes, cf. Figure  4 :

Learning mode (continuous) divided into the following steps:

Import time series from A-Trader to \(H_2O\) platform ( \(H_2O\) is external module of A-Trader, therefore data are imported indirectly from A-Trader database, Notification Agent signals are not used),

Deep Learning (DL) model specification,

DL model Parametrization – (parameters such as;- number of training epochs, number of hidden layers, stopping rounds, stopping metrics, etc),

Building of DL model – on the basis of imported data structure and determined parameters,

Learning and Testing – where the training and the validation datasets are used. Long Short Term Memory architecture of the deep neural network was used. The architecture and hyperparameters of the model are as follows: three hidden LSTM layers (16, 8 and 4 units), dropout layer (rate 0.3), RELU activation function for hidden layers and linear activation function for output layer, loss function: mean squared error, optimizer: adam, metrics: mse, mae, mape, msle, number of epochs:100, batch size:32.

Forecast mode (continuous) – time series of quotations are continuously imported from the A-Trader database, and the trained model is used for predicting rates of return.

The DeepLearning H \(_2O\) Agent is supported also by the following agents [ 43 ]:

Basic Agents - perform time series pre-processing and compute the basic indicators; agents can learn and change their parameters and internal states based on their knowledge.

Intelligent Agents – running on the basis of artificial intelligence (genetic algorithms, rule-based systems, neural networks including MLP, etc.), text messages analysis-based agents, market behavior-based agents.

Decisions of Basic Agents and Intelligent Agents are sent to the Supervisor Agent.

Formally, the model used by DeepLearning H \(_2O\) Agent is defined as follows:

where: \(x^i\) is an input vector of the main quote rate of return, \(x^p,\dots ,x^q\) are inputs vectors consist of the rates of return of the quotations correlated with main quotation (e.g., main quotation is EUR/USD and correlated quotations are gold quotations and oil quotations).

This model uses log-return rates, calculated as follows:

where S \(^{i}_{t}\) denotes a price of quotation i at time t .

\(H_2O\) normalizes log-return rates and projects them in the range from -1 to 1. Input vector related to main quotation is defined as follows:

where k denotes the number of past quotations used as input.

Example of strategy visualization

\(Y_{t+1}\) values are in the range [-1, 1] (generated as fuzzy logic signals) and predict logarithmic return rates at time \(t+1\) (normalized value).

The training set consists of input vectors \(x^i\) and inputs \(x^p,\dots ,x^q\) at time t , \(t-1\) , etc., and output at time \(t+1\) . The learning process is performed on the basis of historical time series; hence the log-return rate at time \(t+1\) is known.

The Supervisor Agent uses different strategies to generate opening/closing positions, on the basis of the output of DeepLearning H \(_2O\) Agent using for instance, consensus strategy or a genetic algorithm, whereas a genetic algorithm determines threshold levels for open/closed short/long positions. The Supervisor also determines the mode of DeepLearning H \(_2O\) Agent operation. If the performance of DeepLearning H \(_2O\) Agent is low (performance measuring issues are presented in the next section), then a learning mode is initiated. If performance is high, a forecasting mode is run using a previously generated model.

The strategies provided by A-Trader can be reused and extended. The user (trader) can add a new agent or source of information by filling out a generic pattern of the agent structure. This is a process of inserting selected agents into your trading strategy.

6 Experiments

The main aim of the experiments is to evaluate the performance of selected trading strategies. The specific aims are as follows:

running the investment strategies, developed in A-Trader, using real data form FOREX market,

the assessment of long/short positions results using return-based and risk-based measures,

comparing the performance of strategies to Buy and Hold benchmark,

confirming the results using statistical tests.

Back-testing is used to verify that the A-Trader strategies were based on the following.

GBP/PLN quotations were selected from randomly selected periods, namely

16-04-2018, 0:00 am to 19-04-2018, 23:59 pm,

23-04-2018, 0:00 am to 26-04-2018, 23:59 pm,

14-05-2018, 0:00 am to 17-05-2018, 23:59 pm.

The strategies MyStrategy , Consensus , Evolution-based were used to generate trading signals (open long/close short position equals 1, close long/open short position equals -1). Figure  5 presents an example (with description) of generated signals (the green line denotes the "long position", the red one denotes the "short position").

The Buy and Hold (B&H) strategy was used as a benchmark (the B&H strategy relies on opening a position at the beginning of the investment period and closing it at the end of this period).

Performance analysis ratios (absolute ratios) were measured in ’pips’ (a change in FOREX price of a ’point’ is called a pip).

The cost of transactions is directly proportional to the number of transactions.

It was assumed that in each transaction the investor engages 100% of the capital held where the trader can individually determine the capital management strategy.

The following measures (ratios) were used in the performance analysis [ 44 , 45 , 46 , 47 , 48 ]:

Rate of return (ratio \(x_1\) ),

Number of transactions,

Gross profit (ratio \(x_2\) ),

Gross loss (ratio \(x_3\) ),

Number of profitable transactions (ratio \(x_4\) ),

Number of profitable consecutive transactions (ratio \(x_5\) ),

Number of unprofitable consecutive transactions (ratio \(x_6\) ),

Sharpe ratio (ratio \(x_7\) ),

Average coefficient of variation (ratio \(x_8\) ),

Average rate of return per transaction (ratio \(x_9\) ), counted as the quotient of the rate of return and the number of transactions.

For comparison of the agent performance, the evaluation function was elaborated, defined as follows:

where \(x_i\) denotes the normalized values of ratios from \(x_1\) to \(x_9\) (mentioned in item 6). For this experiment, coefficients were set as follows: \(x_1\) to \(x_9=1/9\) . However, it is possible to adopt other values for these coefficients. They can be modified using, for example, an evolution-based method, or they can be determined by the trader according to his/her preferences. The functions can be easily modified, and they aggregate many assessment indicators so that users can choose which assessment criteria are most important to them. For example, a trader may be interested in achieving a high rate of return with a high level of risk or a low risk with a low rate of return. Coefficients are needed because the user can arbitrarily classify individual components. The function y returns values from the range \([0\dots 1]\) , and the agent’s performance is assigned proportionally to the function value. This is just one of the evaluation functions, as A-Trader allows a user to build other functions.

Table  1 presents the results of the performance analysis. A wide number of changes in particular ratio values significantly hinder the analysis by the trader and. Consequently, making decisions in time close to real time is very difficult. The results of the experiment allow us to come to the conclusion that the strategy ranking differs in particular periods.

In the first and second periods, Deep learning was the best evaluated strategy. In the third period, the best was the Evolution-based strategy. MyStrategy was evaluated worse than Deep learning and Consensus and B&H was ranked the lowest in all periods.

Considering all periods, it can be stated that the highest rate of return characterized the Deep learning strategy, it was ranked highest in two of the three periods. There was a lower value of the evaluation function in the third period than in Consensus case, which may result from lower values of ratios such as the average rate of return per transaction and risk measures. The Consensus strategy achieved the lowest values for risk measures. It can also be concluded that the low evaluation of MyStrategy in all periods is due not only to the level of the rate of return but also to a high risk level and a large number of unprofitable consecutive transactions. The MyStrategy is simple strategy based on decisions generated by particular agents. The results achieved by MyStrategy allow us to draw conclusions that more sophisticated multi-agent-based methods, such as consensus or deep learning, can perform better than simple strategies. The comparison of multi-agent-based methods and stand-alone methods is presented in our earlier research, for example [ 13 , 36 , 37 ].

The evaluation analysis in other trading systems (e.g., Trade Chimp, XTRADE, MetaTrader) is performed "manually" by the investor in most cases, and this is a very time-consuming process during which there is limited working of the system in real time. These systems offer basic performance measures: rate of return, highest profit, highest loss, number of transactions, total profit, number of profitable transactions, number of profitable consecutive transactions, number of unprofitable consecutive transactions. A-Trader calculates additional ratios, such as risk measurements (average coefficient of variation, Sharpe ratio), or the average rate of return for a specific transaction.

The A-Trader evaluation function enables the measurement and evaluation of investment strategies. These operations are performed automatically by the Supervisor Agent (in time close to real time), which may then advise the investor to trade on the basis of the decisions generated by the strategy characterized by the highest performance level. In addition, users can change the parameters \(a_i\) and \(x_i\) of this function to consider the preferences of the user related to particular performance measures. To confirm the results, statistical tests were performed separately for particular periods using the rate of returns generated by particular transactions in selecting a given strategy as input data. PQStat software  Footnote 2 was used for this and the following hypothesis was assumed:

\(H_0\) – the given strategy was not the best in the given period (the rates of return achieved are not statistically significant).

\(H_1\) – the given strategy is the best in the given period (the rates of return achieved are statistically significant).

First, normality tests were performed. Data are characterized by a nonnormal distribution at the 5% significance level; therefore, a Friedman ANOVA test was performed that included POST-HOC (Dunn Bonferroni). The results are presented in Tables  2 ,  3 , and  4 .

The calculated p-values between the returns rates generated by particular strategies are less than 0.05 in all periods. The lower probability of the p-value indicates stronger evidence against the null hypothesis. Therefore, the null hypothesis can be rejected and the return rates generated by all strategies are statistically significant, suggesting that there is a significant difference between strategies. By ranking the strategies according to the performance scores on three series of quotes, the Deep Learning strategy can be rated the highest.

7 Conclusions

The paper delves into several crucial aspects of designing decision support systems for stock traders through the lens of a multi-agent platform. Within the presented A-Trader system, agents autonomously generate buy-sell decisions using various methods and algorithms, which serve as the foundation for crafting investment strategies. Given the diversity of these decisions and strategies, the evaluation process is overseen by a specialized program known as the Supervisor Agent. This agent enables autonomous selection of the most suitable strategy in near-real time, determining when to open or close long and short positions based on the best strategy identified for a given period. The results of the experiments described in this paper and previous experiments (see [ 36 , 49 ]) highlight that the performance of specific decisions or strategies fluctuates in response to the prevailing conditions in the FOREX market. Through many experiments, it has been clearly demonstrated that no single agent or strategy consistently outperforms others across all periods. The introduction of an evaluation function further enhances this process. A-Trader distinguishes itself with its remarkable flexibility in configuring variables and evaluation functions, providing a dynamic, data-driven platform for user engagement. Investors can assess various strategies regarding returns and risks, allowing for tailored adjustments aligned with their unique requirements. In addition, A-Trader encompasses a broad spectrum of performance measures, including risk-focused metrics, underscoring the critical role of risk management. This emphasis is rooted in the inherent uncertainty and risk associated with financial investments in the FOREX market, influenced by economic cycles, interest rates, government policies, and exchange rates [ 53 ]. In contrast to existing platforms, A-Trader harnesses the consensual advice generated by multiple software agents that are proficient in fundamental, technical, and behavioral analyses [ 18 ]. Crucially, A-Trader refrains from imposing uniform evaluation strategies or functions on every user. The construction of the investment strategy assessment function remains an open endeavor, acknowledging that a one-size-fits-all solution may not exist. The paper effectively illustrates that the suitability of a linear function is expected. However, adopting a non-linear function can be intricate and must be more readily understandable to investors, often shrouded in secrecy among financial experts. A-Trader stands as an open system, giving users the tools to fashion their strategies and seamlessly integrate strategies created in other software environments.

This research offers several noteworthy contributions both to the scientific understanding of financial decision support systems and the practical application of these systems in real-world trading scenarios, notably in the areas of:

Integration of Diverse Decision Sources: A-Trader integrates a wide range of decision sources by enabling multiple agents to generate independent buy-sell decisions. This diversity of sources provides a comprehensive view of market dynamics and contributes to a more holistic decision-making process.

Agent supervision: The introduction of the Supervisor Agent serves as a pivotal contribution. This agent takes on the task of evaluating the heterogeneity of decisions and strategies. It intelligently selects the best strategy in response to the current market conditions, offering traders a pragmatic solution.

Dynamic Strategy Selection: The research highlights that no single agent or strategy consistently outperforms others in all market conditions. This observation underscores the need for an adaptive and dynamic approach to strategy selection. Using an evaluation function empowers the Supervisor Agent to automatically identify the best strategy in near-real time, enhancing investment effectiveness and responsiveness to market changes.

Flexibility in Evaluation Functions: A-Trader allows users to configure variables and evaluation functions, promoting a data-driven approach to user engagement. This flexibility ensures that investors can tailor their strategies based on their unique risk tolerance and performance criteria preferences.

Risk Management Integration: A-Trader acknowledges the inherent risk and uncertainty associated with financial investments in the FOREX market. By considering a wide range of performance measures, including risk-based metrics, the platform emphasizes the importance of risk management. This is a crucial scientific contribution, as it addresses a key challenge in real-world trading.

Open System and Interoperability: A-Trader’s open system architecture is a scientific breakthrough. It allows users to seamlessly build their strategies and integrate strategies from other software environments. This interoperability enhances the practical utility of the platform, making it adaptable to the diverse needs of traders and investors.

In a pragmatic sense, A-Trader offers traders, investors, and market participants a sophisticated tool that leverages multiple agents for decision support. Provides a more adaptable and responsive approach to trading in the dynamic FOREX market. In addition, it is a pioneering platform that bridges the gap between scientific research and practical trading strategies. The limitation of this approach is the high computational complexity it entails. For example, when A-Trader runs for a month, it processes a substantial amount of data, approximately 1TB. In addition, there is a lack of direct communication between agents, and the Notification Agent acts as an intermediary to transmit signals. Consequently, this agent is a critical component of system performance. The limitation of this research is that we used only one pair of quotations in the experiments. These challenges will be the focus of further research. The ongoing research will include developing a directional change algorithm , an evolutionary approach to determine learning parameters, and implementing cognitive agents based on fundamental analysis and expert opinions. Further research on the application of spiking neural networks in a-Trader should also be performed. Overall, the contributions of this work extend beyond the theoretical realm, demonstrating a commitment to addressing the practical challenges traders and investors face in real-time decision making within financial markets. This multidimensional approach to financial decision support promises to enhance investment effectiveness and contribute to the broader field of algorithmic trading.

Data Availability

Data will be made available on request.

W3C SOAP , https://www.w3.org/TR/soap/

PQStat software ,  https://pqstat.pl/

Rundo F (2019) Deep lstm with reinforcement learning layer for financial trend prediction in fx high frequency trading systems. Appl Sci 9(20). https://doi.org/10.3390/app9204460

Chen HJ, Chen SJ, Chen Z, Li F (2019) Empirical investigation of an equity pairs trading strategy. Manage Sci 65(1):370–389

Article   Google Scholar  

Tadelis S (2016) Reputation and feedback systems in online platform markets. Annual Review of Economics 8(1):321–340. https://doi.org/10.1146/annurev-economics-080315-015325

Freitas FD, Freitas CD, De Souza AF (2013) System architecture for on-line optimization of automated trading strategies. In: Proceedings of the 6th workshop on high performance computational finance. WHPCF ’13. Association for Computing Machinery, New York, NY, USA. https://doi.org/10.1145/2535557.2535563

Zhou S (2023) Forex automated trading system establishment and optimization analysis. Inform Syst Econ 63–69. https://doi.org/0.23977/infse.2023.040710

Roostaee MR, Abin, AA (2023) Forecasting financial signal for automated trading: An interpretable approach. Expert Syst Appl 211:118570. https://doi.org/10.1016/j.eswa.2022.118570

Ismail MAH, Yasruddin ML, Husin Z, Tan WK (2022) Automated trading system for forecasting the foreign exchange market using technical analysis indicators and artificial neural network. In: 2022 IEEE 18th International colloquium on signal processing & applications (CSPA), pp 63–68. https://doi.org/10.1109/CSPA55076.2022.9781856

Aru O, Okechukwu C (2023) development of an optimized intelligent machine learning approach in forex trading using moving average indicators. LAUTECH Journal of Engineering and Technology 17(2):18–27

Google Scholar  

Thompson JR (2013) Analysis of market returns using multifractal time series and agent-based simulation. PhD thesis. AAI3575853

Wah E, Wellman MP (2013) Latency arbitrage, market fragmentation, and efficiency: A two-market model. In: Proceedings of the fourteenth ACM conference on electronic commerce. EC ’13, pp. 855–872. Association for Computing Machinery, New York, NY, USA. https://doi.org/10.1145/2482540.2482577

Ishikawa K, Nakata K (2021) Online trading models with deep reinforcement learning in the forex market considering transaction costs. arXiv:2106.03035

Taleb NN (2018) Election predictions as martingales: an arbitrage approach. Quantitative Finance 18(1):1–5. https://doi.org/10.1080/14697688.2017.1395230

Article   MathSciNet   Google Scholar  

Korczak J, Hernes M, Bac M (2017) Collective intelligence supporting trading decisions on forex market. In: Nguyen NT, Papadopoulos GA, Jędrzejowicz P, Trawiński B, Vossen G (eds) Computational Collective Intelligence. Springer, Berlin, Heidelberg, pp 113–122

Chapter   Google Scholar  

Huang Z, Tanaka F (2022) Mspm: A modularized and scalable multi-agent reinforcement learning-based system for financial portfolio management. PLoS ONE 17(2):1–24. https://doi.org/10.1371/journal.pone.0263689

Shavandi A, Khedmati, M (2022) A multi-agent deep reinforcement learning framework for algorithmic trading in financial markets. Expert Syst Appl 208:118124. https://doi.org/10.1016/j.eswa.2022.118124

He F-F, Chen C-T, Huang S-H (2023) A multi-agent virtual market model for generalization in reinforcement learning based trading strategies. Appl Soft Comput 134:109985. https://doi.org/10.1016/j.asoc.2023.109985

Vadori N, Ardon L, Ganesh S, Spooner T, Amrouni S, Vann J, Xu M, Zheng Z, Balch T, Veloso M (2024) Towards multi-agent reinforcement learning-driven over-the-counter market simulations. Math Financ 34(2):262–347. https://doi.org/10.1111/mafi.12416

Loginov A, Heywood MI (2014) On evolving multi-agent fx traders. In: Esparcia-Alcázar AI, Mora AM (eds) Applications of Evolutionary Computation. Springer, Berlin, Heidelberg, pp 203–214

Loginov A, Heywood MI (2013) On the utility of trading criteria based retraining in forex markets. In: Esparcia-Alcázar AI (ed) Applications of evolutionary computation. Springer, Berlin, Heidelberg, pp 192–202

Kahneman D, Rosenfield AM, Gandhi L, Blaser T (2016) Reducing noise in decision making. Harv Bus Rev 94(12):18

Kočišová J, Horváth D, Kasanický T, Buša J (2012) Prediction of financial markets using agent-based modeling with optimization driven by statistical evaluation of historical data. In: Adam G, Buša J, Hnatič M (eds) Mathematical modeling and computational science. Springer, Berlin, Heidelberg, pp 308–313

Sarani D, Rashidi-Khazaee, DP (2024) A Deep Reinforcement Learning Approach for Trading Optimization in the Forex Market with Multi-Agent Asynchronous Distribution. https://arxiv.org/abs/2405.19982

Abraham A (2002) Analysis of hybrid soft and hard computing techniques for forex monitoring systems. In: 2002 IEEE World congress on computational intelligence. 2002 IEEE International conference on fuzzy systems. FUZZ-IEEE’02. Proceedings, vol 2, pp 1616–1621. https://doi.org/10.1109/FUZZ.2002.1006749

Sher GI (2012) Forex trading using geometry sensitive neural networks. In: Proceedings of the 14th annual conference companion on genetic and evolutionary computation. GECCO ’12, pp 1533–1534. ACM, New York. https://doi.org/10.1145/2330784.2331032

Khosravi H, Shiri ME, Khosravi H, Iranmanesh E, Davoodi A (2009) Tactic- a multi behavioral agent for trading agent competition. In: Sarbazi-Azad H, Parhami B, Miremadi S-G, Hessabi S (eds) Advances in computer science and engineering. Springer, Berlin, Heidelberg, pp 811–815

Mozetic I, Gabrovsek P, Novak PK (2018) Forex trading and twitter: Spam, bots, and reputation manipulation. CoRR abs/1804.02233

Hinton GE, Osindero S, Teh Y-W (2006) A fast learning algorithm for deep belief nets. Neural Comput 18(7):1527–1554. https://doi.org/10.1162/neco.2006.18.7.1527

Kuremoto T, Kimura S, Kobayashi K, Obayashi M (2014) Time series forecasting using a deep belief network with restricted boltzmann machines. Neurocomputing 137:47–56. https://doi.org/10.1016/j.neucom.2013.03.047

Yang S, Chen B (2023) Snib: Improving spike-based machine learning using nonlinear information bottleneck. IEEE Transactions on Systems, Man, and Cybernetics: Systems 53(12):7852–7863. https://doi.org/10.1109/TSMC.2023.3300318

Yang CBS (2023) Effective surrogate gradient learning with high-order information bottleneck for spike-based machine intelligence. IEEE Transactions on neural networks and learning systems. https://doi.org/10.1109/TNNLS.2023.3329525

Yang S, Pang Y, Wang H, Lei T, Pan J, Wang J, Jin Y (2023) Spike-driven multi-scale learning with hybrid mechanisms of spiking dendrites. Neurocomputing 542:126240. https://doi.org/10.1016/j.neucom.2023.126240

Gerlein EA, McGinnity M, Belatreche A, Coleman S (2016) Evaluating machine learning classification for financial trading: An empirical approach. Expert Syst Appl 54:193–207. https://doi.org/10.1016/j.eswa.2016.01.018

Talebi H, Hoang W, Gavrilova ML (2014) Multi-scale foreign exchange rates ensemble for classification of trends in forex market. Procedia Computer Science 29:2065–2075. https://doi.org/10.1016/j.procs.2014.05.190

Mayo M (2012) Evolutionary data selection for enhancing models of intraday forex time series. In: Di Chio C at al (eds) Applications of evolutionary computation, Springer, Berlin, Heidelberg, pp. 184–193

Agarwal S (2011) Toward a push-scalable global internet. In: 2011 IEEE Conference on computer communications workshops (INFOCOM WKSHPS), pp 786–791. https://doi.org/10.1109/INFCOMW.2011.5928918

Korczak J, Hernes M, Bac M (2013) Risk avoiding strategy in multi-agent trading system. In: Ganzha M, Maciaszek MPL (ed.) Proceedings of the 2013 federated conference on computer science and information systems, IEEE, Los Alamitos, CA, pp 1119–1126

Korczak J, Hernes M, Bac M (2014) Performance evaluation of decision-making agents’ in the multi-agent system. In: 2014 Federated Conference on Computer Science and Information Systems, pp 1171–1180. https://doi.org/10.15439/2014F188

Longo JM (2014) Trading and Investment Strategies in Behavioral Finance, Wiley Ltd, Hoboken, New Jersey. Chap. 27, pp 495–512. https://doi.org/10.1002/9781118813454.ch27

Shiller RJ (2012) Finance and the Good Society. Princeton University Press, New Jersey, United States

Lento C (2009) The combined signal approach to technical analysis: A review & commentary. SSRN Electron J. https://doi.org/10.2139/ssrn.1410899

Hernes M, Nguyen NT (2007) Deriving consensus for hierarchical incomplete ordered partitions and coverings. J Univ Comput Sci 13(2):317–328

Hernes M, Sobieska-Karpińska J (2016) Application of the consensus method in a multiagent financial decision support system. Inf Syst E-bus Manag 14(1):167–185. https://doi.org/10.1007/s10257-015-0280-9

Korczak J, Hernes M (2017) Deep learning for financial time series forecasting in a-trader system. In: 2017 Federated conference on computer science and information systems (FedCSIS), pp 905–912. https://doi.org/10.15439/2017F449

Hu D, Schwabe G, Li X (2015) Systemic risk management and investment analysis with financial network analytics: research opportunities and challenges. Financ Innov 1(1):2. https://doi.org/10.1186/s40854-015-0001-x

Hussain OK, Dillon TS, Hussain FK, Chang EJ (2013) Risk Assessment Phase: Financial Risk Assessment in Business Activities, pp 151–185. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-28690-2_6

Lückoff P (2011) Mutual Fund Performance and Performance Persistence. Gabler Verlag, Wiesbaden. https://doi.org/10.1007/978-3-8349-6527-1

Qiu Z (2016) Discussion of investment analysis method in the new round of the china stock bull market. In: Li M, Zhang Q, Zhang J, Li Y (eds) Proceedings of 2015 2nd international conference on industrial economics system and industrial security engineering, Springer, Singapore, pp 311–317

Yao Y-y, Zhang R-s (2016) Empirical research on efficiency measure of financial investment in education based on se-dea. In: Cao B-Y, Liu Z-L, Zhong Y-B, Mi H-H (eds) Fuzzy systems & operations research and management, Springer, Cham, pp 389–402

Korczak J, Hernes M (2018) Performance evaluation of trading strategies in multi-agent systems - case of a-trader. In: 2018 Federated conference on computer science and information systems (FedCSIS), pp 839–844

Mancas D, Udristoiu S, Manole E, Lapadat B (2008) A comparison of multi-agents competing for trading agents competition. WSEAS TRANS- ACTIONS on COMPUTERS 7(12):1916–1926

Calzi ML, Milone L, Pellizzari P (2010) Progress in Artificial Economics: Computational and Agent-Based Models. Springer, Berlin, Heidelberg

Eiben AE, Smith JE (2015) Introduction to Evolutionary Computing, 2nd edn. Springer, Berlin, Heidelberg, Natural Computing Series. https://doi.org/10.1007/978-3-662-44874-8

Book   Google Scholar  

LeBaron B (2011) Active and passive learning in agent-based financial markets. Eastern Economic Journal 37(1):35–43. https://doi.org/10.1057/eej.2010.53

Download references

This research was founded by the Ministry of Science and Higher Education in Poland under the program "Regional Initiative of Excellence" [No. 015/RID/2018/19].

Author information

Authors and affiliations.

Department of Process Management, Wrocław University of Economics, Komandorska 118/120, Wrocław, 53-345, Poland

Marcin Hernes & Maciej Pondel

International University of Logistics and Transport in Wroclaw, Sołtysowicka 19B, Wrocław, 51-168, Poland

Jerzy Korczak

Department of Applied Informatics, Wrocław University of Science and Technology, Wyb. Wyspiańskiego 27, Wroc,ław, 50-370, Poland

Dariusz Krol

Department of Information Systems, University of Münster, Leonardo Campus 3, Münster, 9, Germany

Jörg Becker

You can also search for this author in PubMed   Google Scholar

Contributions

Conceptualization: Jerzy Korczak, Marcin Hernes; Methodology: Jerzy Korczak, Marcin Hernes; Formal analysis and investigation: Jörg Becker, Dariusz Król; Writing - original draft preparation: Marcin Hernes, Maciej Pondel, Dariusz Król; Writing - review and editing: Jerzy Korczak, Jörg Becker; Funding acquisition: Marcin Hernes; Resources: Marcin Hernes, Maciej Pondel; Supervision: Jerzy Korczak, Marcin Hernes.

Corresponding author

Correspondence to Marcin Hernes .

Ethics declarations

Competing of interest.

The authors have no competing interests to declare that are relevant to the content of this article.

Ethical and informed consent for data used

This article does not involve any studies with human participants or animals performed by any of the authors.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Hernes, M., Korczak, J., Krol, D. et al. Multi-agent platform to support trading decisions in the FOREX market. Appl Intell (2024). https://doi.org/10.1007/s10489-024-05770-x

Download citation

Accepted : 11 August 2024

Published : 30 August 2024

DOI : https://doi.org/10.1007/s10489-024-05770-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Multi-agent system
• Supporting trading decision-making
• FOREX market
• Find a journal
• Publish with us
• Track your research

Main Navigation

• Contact NeurIPS
• Code of Ethics
• Code of Conduct
• Create Profile
• Journal To Conference Track
• Diversity & Inclusion
• Proceedings
• Future Meetings
• Exhibitor Information
• Privacy Policy

NeurIPS 2024, the Thirty-eighth Annual Conference on Neural Information Processing Systems, will be held at the Vancouver Convention Center

Monday Dec 9 through Sunday Dec 15. Monday is an industry expo.

Registration

Pricing » Registration 2024 Registration Cancellation Policy »

Career Site Conference Hotels NeurIPS has contracted Hotel guest rooms for the Conference at group pricing, requiring reservations only through this page. Please do not make room reservations through any other channel, as it only impedes us from putting on the best Conference for you. We thank you for your assistance in helping us protect the NeurIPS conference.

Announcements

• Financial Grant and Volunteer applications are open 
• The 2024 visa process has changed, see the Visa Information page for details.
• The  accepted competitions  have been released.

Latest NeurIPS Blog Entries [ All Entries ]

Important Dates

If you have questions about supporting the conference, please contact us .

View NeurIPS 2024 exhibitors » Become an 2024 Exhibitor Exhibitor Info »

Organizing Committee

General chair, program chair, workshop chair, workshop chair assistant, tutorial chair, competition chair, data and benchmark chair, diversity, inclusion and accessibility chair, affinity chair, ethics review chair, communication chair, social chair, journal chair, creative ai chair, workflow manager, logistics and it, mission statement.

The Neural Information Processing Systems Foundation is a non-profit corporation whose purpose is to foster the exchange of research advances in Artificial Intelligence and Machine Learning, principally by hosting an annual interdisciplinary academic conference with the highest ethical standards for a diverse and inclusive community.

About the Conference

The conference was founded in 1987 and is now a multi-track interdisciplinary annual meeting that includes invited talks, demonstrations, symposia, and oral and poster presentations of refereed papers. Along with the conference is a professional exposition focusing on machine learning in practice, a series of tutorials, and topical workshops that provide a less formal setting for the exchange of ideas.

More about the Neural Information Processing Systems foundation »