Pictures copyrighted

H. La Poutré - Adaptive Logistics in Hospitals: Computational Intelligence Techniques and Agent-based Simulations

In recent years, the need for efficient and effective planning in hospitals has become prominent. Increasing numbers of patients as well as an aging population has increased the demand for health care. In order to limit or even reduce the costs for health care, efficient planning methods need to be developed. In addition, hospitals are changing their view on health care processes and nowadays like to deliver their services in a patient-oriented way: Health care services should answer to the patient's preferences, in a comfortable way.

Health care processes are characterized by a high level of uncertainty. Patients can have complications, in-patients may need to stay longer than expected, and urgencies may occur. In addition, the health care organisation is highly decentralized: hospitals consist of autonomous medical departments, that together have to treat a patient. Thus, health care logistics has to deal with uncertainty and decentralisation.

In the presentation, we present some solutions for health care logistics problems. In order to deal with the uncertainty and decentralisation, we show planning solutions by means of techniques from multi-agent systems and computational intelligence. In particular, we address resource management in hospitals for in-patients. We present techniques for dynamic planning and adaptation of the number of beds in various departments (like Intensive Care, Medium Care, etc.) for in-patients having a surgery. Also, we consider patient appointment planning for out-patients in hospitals, where urgency and preferences have to be taken into account, next to the efficiency of the planning methods. The methods and techniques have been developed in cooperations between CWI, TU Eindhoven, and the hospitals AMC (Amsterdam) and CZE (Eindhoven).

(Joint work with: Anke Hutzschenreuter, Ivan Vermeulen, Peter Bosman, and Sander Bohte.)