Scheduling decisions in hospitals are often taken in a decentralized way. This means that different specialized hospital units decide autonomously on e.g. patient admissions and schedules of shared resources. Decision support in such a setting requires methods and techniques that are different from the majority of existing literature in which centralized models are assumed. The design and analysis of such methods and techniques is the focus of this thesis. Specifically, we develop computational models to provide dynamic decision support for hospital resource management, the prediction of future resource occupancy and the application thereof. Hospital resource management targets the efficient deployment of resources like operating rooms and beds. 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. The issues are further complicated by the fact that patient arrivals are dynamic and treatment processes are stochastic. Our approach to providing decision support combines techniques from multi-agent systems and computational intelligence (CI). This combination of techniques allows to properly consider the dynamics of the problem while reflecting the distributed decision making practice in hospitals. Multi-agent techniques are used to model multiple hospital care units and their decision policies, multiple patient groups with stochastic treatment processes and uncertain resource availability due to overlapping patient treatment processes. The agent-based model closely resembles the real-world situation. Optimization and learning techniques from CI allow for designing and evaluating improved (adaptive) decision policies for the agent-based model, which can then be implemented easily in hospital practice. In order to gain insight into the functioning of this complex and dynamic problem setting, we developed an agent-based model for the hospital care units with their patients. To assess the applicability of this agent-based model, we developed an extensive simulation. Several experiments demonstrate the functionality of the simulation and show that it is an accurate representation of the real world. The simulation is used to study decision support in resource management and patient admission control. To further improve the quality of decision support, we study the prediction of future hospital resource usage. Using prediction, the future impact of taking a certain decision can be taken into account. In the problem setting at hand for instance, predicting the resource utilization resulting from an admission decision is important to prevent future bottlenecks that may cause the blocking of patient flow and increase patient waiting times. The methods we investigate for the task of prediction are forward simulation and supervised learning using neural networks. In an extensive analysis we study the underlying probability distributions of resource occupancy and investigate, by stochastic techniques, how to obtain accurate and precise prediction outcomes. To optimize resource allocation decisions we consider multiple criteria that are important in the hospital problem setting. We use three conflicting objectives in the optimization: maximal patient throughput, minimal resource costs and minimal usage of back-up capacity. All criteria can be taken into account by finding decision policies that have the best trade-off between the criteria. We derived various decision policies that partly allow for adaptive resource allocations. The design of the policies allows the policies to be easily understandable for hospital experts. Moreover, we present a bed exchange mechanism that enables a realistic implementation of these adaptive policies in practice. In our optimization approach, the parameters of the different decision policies are determined using a multiobjective evolutionary algorithm (MOEA). Specifically, the MOEA optimizes the output of the simulation (i.e. the three optimization criteria) as a function of the policy parameters. Our results on resource management show that the benchmark allocations obtained from a case study are considerably improved by the optimized decision policies. Furthermore, our results show that using adaptive policies can lead to better results and that further improvements may be obtained by integrating prediction into a decision policy.
|Qualification||Doctor of Philosophy|
|Award date||26 May 2010|
|Place of Publication||Eindhoven|
|Publication status||Published - 2010|