460 / 2024-04-25 15:36:49

Integrated Planning and Scheduling of Operating Rooms for Elective Surgeries with Time Window Preferences

Operating Room Planning and Scheduling, Elective Surgeries, Time Windows Preference, Stochastic Surgery Times

Motivated by common practices in hospital operations, we study integrated planning

and scheduling of operating rooms (ORs) for elective surgeries over multiple days. The decisions of the problem involve

determining a subset of ORs to open on each day, assigning the available surgeries to specific days and ORs, and schedul-

ing the assigned surgeries in each room on each day, in the presence of eligibility constraints between ORs and surgeries,

surgery ready dates and deadlines, and surgery intra-day time window preferences. The objective is to minimize the sum

of the total setup cost of opening ORs, overtime cost of the opened ORs and surgery time window violation penalty cost.

We consider both the deterministic version of the problem where the surgery operating times are

assumed to be certain and known in advance, and the stochastic version where the surgery operating times are random.

To solve these problems, we propose an efficient column generation based heuristic for the deterministic problem and a

Monte Carlo sampling based method for the stochastic problem. In addition, we identify a frequently encountered special

case with a relatively small number of planning days and a relatively small number of surgery types, and propose poly-

nomial time algorithms to solve it to optimality under both deterministic and stochastic settings. Computational results

based on real data demonstrate the effectiveness of the proposed solution approaches. Our

results reveal that the existence of surgery intra-day time windows has a major impact on the total cost of surgery sched-

ule, especially when they are relatively tight. In this case, OR managers should provide incentives to the patients and the

surgeons to adjust surgery time windows. We also demonstrate the impact of surgery duration randomness, and identify

different conditions for the suitability of the deterministic and stochastic mode