Operational cost analysis for e-commerce deliveries using agent-based modeling and simulation

Abstract

The COVID-19 pandemic has highlighted the growth of e-commerce around the world. Especially in developing countries, assisted home delivery (AHD) is the main way for people to receive their goods purchased online. This delivery policy can result in delivery failure when the consumer is not at home and, consequently, in greater externalities that impact all stakeholders in last mile logistics. This paper aims to evaluate the operating costs of delivery lockers (DLs) through agent-based modeling and simulation. A total of 84 scenarios were simulated alternating the DLs usage rate and different delivery attempt policies. The results show that the most impacting system costs are time, external and re-delivery costs. The implementation of DLs and the exclusion of the three-attempt delivery together can reduce these costs and increase net profit by up to 79.1%. Therefore, DLs have economic potential for last mile logistics operations. However, this solution must be accompanied by incentive policies resulting from stakeholder engagement. We hope that this study contributes to a greater understanding of DLs as a sustainable solution for last mile e-commerce deliveries, helping decision makers in urban freight transport.