Online optimization methods applied to the management of health services
Online optimization methods applied to the management of health services.
Abstract: The use of Operations Research in health care delivery has developed considerably over the years. Many decision problems regarding the management of health services deal with unpredictable demand, events or variables, which make them more challenging. We can address such an issue with Online Optimization, which is characterized by the development of algorithms whose decisions are based only on partial information that becomes available over time. Online optimization takes into account the partial information obtained from the past and exploits the concept of lookahead, that is a limited overseen amount of future input data: such information can be derived by the knowledge of the Clinical Pathway (CP) or through predictive approaches. In this talk, we propose Online Optimization approaches for problems arising in two different CPs: the real-time management of the operating rooms (in the context of a well-structured process) and the dynamic resource allocation at the Emergency Department (which deals with a so-called spaghetti-process). While algorithms can exploit the solid knowledge of the CP for the former, a predictive model based on process mining is proposed for the latter in order to compensate for the lack of information about the possible patient path evolution.