A variable neighborhood search algorithm for an integrated physician planning and scheduling problem

In this paper, we study an integrated physician planning and scheduling problem (IPPSP) considering multiple types of services, which can be done by multiple qualifications of physicians with different costs and revenues. The problem of physician planning is to assign the working periods to each physician, based on which the problem of physician scheduling is to determine the service type and the number of each type of service to be provided by each physician. Motivated by the increasing financial pressure in most public hospitals, the objective of the problem is to maximize the total revenues by the consultation services minus the costs by the physicians of the outpatient department. An integer programming model is proposed for the IPPSP and we prove the proposed problem to be NP-hard. A Variable Neighborhood Search (VNS) algorithm with an adaptive heuristic with a Dynamic Programming (DP) algorithm embedded is proposed which is named IPPSP-VNS to obtain high-quality solutions in an appropriate time. We compare the algorithm with the commercial solver Gurobi and four state-of-the-art meta-heuristic algorithms. The experimental results show that the proposed IPPSP-VNS can obtain high -quality results and outperform other compared algorithms on solution superiority and efficiency.

Automated summary

This paper studies an integrated physician planning and scheduling problem, proposes an integer programming model, and uses a Variable Neighborhood Search algorithm to obtain high-quality solutions. The algorithm performs superiorly in experiments, outperforming other compared algorithms.