Two-stage deployment and operation plan of PCR sample collection booths under demand uncertainty

Abstract

Polymerase chain reaction (PCR) testing programs have been crucial in combating infectious diseases, such as COVID-19. However, the widespread implementation of frequent nucleic acid testing requires the dense deployment of PCR sample collection booths (SCBs). In this study, we propose two-stage stochastic modeling frameworks to optimize the deployment and operation of SCBs, considering demand uncertainty and both regular and ad hoc services. The first stage optimizes the location and regular staff allocation to achieve a specified service reliability (SR). The second stage determines citizens’ equilibrium SCB choice and allocates ad hoc staff when demand exceeds the capacity determined in the first stage. We propose an SR-based gradient solution approach to efficiently solve the model. Numerical experiments using real-world PCR data demonstrate the effectiveness and efficiency of our framework. Compared with the deployment plan implemented in Shanghai during the pandemic, the optimized deployment and operation plan results in 7% and 12% saving in total cost for the Laoximen subdistrict and the Huangpu district, respectively. The modeling framework and findings presented in this study can inform government decision-making regarding the deployment and operation plan of SCBs in combatting infectious diseases, which can specify the optimal number and location of SCBs to deploy and the regular and ad hoc staff allocation for each SCB to meet the demand at the lowest cost.