A combinatorial auction mechanism for surgical scheduling considering surgeon's private availability information

摘要

This paper addresses surgical scheduling problem in a decentralized environment where the availability of surgeons is considered as their private information. In addition to combinatorial complexities inherited from centralized surgical scheduling models, strategic behaviors of surgeons derived from decentralized game theoretic environments have to be addressed. We propose an iterative auction mechanism, a form of decentralized combinatorial optimization, for solving the surgical scheduling problem. The eligibility restriction of operation rooms is also considered. The objective is to maximize the overall social welfare of patients which is represented by the overall weights of surgeries scheduled. Under the proposed mechanism, we prescribe strategies for surgeons on submitting their availability information to maximize their preference values, and at the same time, minimize the revelation of their private information. We prove that myopic bidding strategy is a weakly dominant strategy for surgeons under the proposed scheduling mechanism and the solution quality is an non-decreasing function of the number of bidding rounds along the bidding process. We also present a nontrivial worked example to illustrate the application of the proposed approach in surgical scheduling setting.