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.