Authors:
Serene Almomen
and
Daniel A. Menascé
Affiliation:
George Mason University, United States
Keyword(s):
Autonomic computing, Healthcare, Emergency department, Quality of service, Length of stay, Self-managed system, Simulation, Combinatorial search.
Related
Ontology
Subjects/Areas/Topics:
Biomedical Engineering
;
Decision Support Systems
;
Design and Development Methodologies for Healthcare IT
;
Health Information Systems
;
Healthcare Management Systems
;
Pattern Recognition and Machine Learning
Abstract:
Improving the operational efficiency of an Emergency Department (ED) is a complex task due the very large number of ED configurations that involve human and physical resources and due to the unpredictable nature of the ED’s workload. Thus, managing the performance of EDs becomes difficult and expensive when carried out by human beings alone. A new approach, called self-managed EDs, consists of building into the ED the mechanisms required to self-adjust the ED’s configuration parameters so that its Quality of Service (QoS) is constantly met. This paper presents the design of an autonomic controller for a self-managed Emergency Department and evaluates its effectiveness in maximizing a utility function of the ED subject to cost-constraints. The controller uses a combination of combinatorial search techniques with simulation models. Experimental results illustrate the operation of the controller and how it reacts to variations of patient interarrival times.