The Urban Recharging Infrastructure Design Problem with Stochastic Demands and Capacitated Recharging Stations

In this study, we introduce the urban recharging infrastructure design problem with stochastic demands and capacitated recharging stations. Stochastic recharge demands shaped by daily transportation/logistics activity cycles, capacity limitations of recharging stations and users’ route preferences need to be considered simultaneously to address this urgent problem that focus on the inner city transportation and logistics use of electric vehicles. We present scenario based stochastic programming formulations to find infrastructure designs that minimize the total investment cost while guaranteeing a target quality of service level that is crucial to facilitate widespread use of electric vehicles in logistics activities. We propose a branch and cut algorithm that can efficiently solve realistic problem instances with over 390,000 nodes, 1,200,000 edges, 76,000 recharging demand configurations, 200 candidate recharging station locations and 1,000 scenarios. Our computational experiments on different network topologies and recharging demand structures attest to the efficiency of our solution approach and present significant managerial insights for the decision makers in both business and government. In particular our results show that high quality of service levels can be achieved with a minor increase in the total investment cost and the multi period investment strategies can be viable alternatives to deal with the uncertainty related to the future demand levels. Recharging Station Location, Urban Logistics, Green Transportation, Infrastructure Investment, Integer Programming, Branch and Cut