Locate vaccination stations considering travel distance, operational cost, and work schedule

In this study, a multi-objective mixed-integer nonlinear programming (MINLP) model is proposed to help the Center for Disease Control and Prevention (CDC) determine the locations of vaccination stations while, simultaneously, considering travel distance, operational cost, and work schedule. A case based on Nanshan CDC in Shenzhen City of China is also studied. Due to the computational complexity, a two-stage strategy is developed to simplify the MINLP into a mixed-integer linear programming (MILP) model, and then ε-constraint method is applied to handle the multiple conflicting objectives. Our research reveals interesting and valuable results. For a given bound of operational cost, the number of fully open stations can greatly increase by wisely allocating demand groups and medical professionals. Although bearing more cost is in favor of decreasing travel distance, it is not beneficial to invest too much due to the decreasing effect, especially when requiring a high number of fully open stations. Current service capacity is large enough even if the birth rate increases by 50%, regarding the “two-child” policy. The case study validates that the proposed model can provide good Pareto non-inferior solutions for the CDC to balance different objectives and show great potential in locating public healthcare resources. Copyright © 2020 Elsevier Ltd. All rights reserved.