Evolutionary Multiobjective Optimization for Dynamic Hospital Resource Management

Allocating resources to hospital units is a major managerial issue as the relationship between resources, utilization and patient flow of different patient groups is complex. Furthermore, the problem is dynamic as patient arrival and treatment processes are stochastic. In this paper we present a strategy optimization approach where the parameters of different strategies are optimized using a multiobjective EDA. The strategies were designed such that they enable dynamic resource allocation with an offline EDA. Also, the solutions are understandable to health care professionals. We show that these techniques can be applied to this real-world problem. The results are compared to allocation strategies used in hospital practice.