Surgery scheduling with elective and emergency demand

Operating rooms (ORs) are one of the most crucial and scarce medical resources. Generally, the elective (EL) patient consults with their physician or surgeon and is usually planned several days ahead of their surgery date. On the contrary, an emergency (EM) surgery has to be performed as soon as possible in order to save human lives. These EM patients could cause EL surgery postponement and aggravate OR congestion, which has become an increasingly severe global public health problem. Especially in mass casualty incidents, such as large accidents and disasters, a hospital may accept too many EM surgeries that exceed the OR capacity. Then, the unpredictable uncertainty and the manual empirical schedule could further cause EL surgery cancellations and delay EM surgeries. How to maximize the demand of EM surgeries accepted by the hospital while adjusting the regular hospital operations as little as possible is a practical problem. To solve this problem, we study the following optimization issues about the surgery scheduling with EL and EM demand: (1) How to plan and schedule EM surgeries? (2) How to reschedule EL surgeries and EM surgeries in real-time based on the current policy? (3) How to schedule next-day EL surgeries considering potential EM demand? These are three successive management topics under uncertainty. We aim at employing Machine Learning, Operations Research, and Simulation techniques to optimize decisions to support the OR managers. Furthermore, we work toward publishing them in the top Management Science journals and applying them to real hospitals within four years.