Agent-oriented Decision Support System for Business Processes Management with Genetic Algorithm Optimization: an Application in Healthcare

• Published in: Journal of medical systems Vol. 44; no. 9; p. 157
• Main Authors: Sulis, Emilio, Terna, Pietro, Di Leva, Antonio, Boella, Guido, Boccuzzi, Adriana
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2020; Springer Nature B.V
• Subjects: Agents (artificial intelligence); Algorithms; Artificial intelligence; Business process management; Computer simulation; Decision support systems; Education & Training; Emergency medical care; Emergency medical services; Environment models; Genetic algorithms; Health care; Health Informatics; Health Sciences; Healthcare Intelligent Multi-Agent Systems (HIMAS2020); Medicine; Medicine & Public Health; Multiagent systems; Optimization; Patients; Statistics for Life Sciences; Three dimensional models; Business process management; Agent-based decision support system; Genetic algorithm; Emergency department planning
• ISSN: 0148-5598; 1573-689X; 1573-689X
• DOI: 10.1007/s10916-020-01608-4

Agent-based approaches have been known to be appropriate as systems and methods in medical administration in recent years. The increased attention to processes led to the recent growth of Business Process Management discipline, which quite exclusively adopt discrete-event modeling and simulation. This paper proposes a medical agent-oriented decision support system to integrate the achievements from management science, agent-based modeling, and artificial intelligence. In particular, we performed a practical application concerning a hospital emergency department medical system. We adopt the widely used multi-agent programmable modeling environment NetLogo. First, we demonstrated the ability to perform a clear representation of healthcare processes where agents (i.e., patients and hospital staff) operate in a 3D environment. This model allows performing a traditional what-if scenario analysis. Second, we explore how performing intelligent management of patients by applying genetic algorithms to find the criteria for the selection process of the subjects in the admission procedure. The results are encouraging towards a more extensive application of agent-oriented methodologies in healthcare management.