Stochastic programming for optimizing sustainable urban logistics: multi-period production-routing and reverse logistic

The increasing need for sustainable logistics and efficient urban transportation systems has driven significant research in integrated optimization models. This thesis addresses the development of multi-period production-routing models tailored to enhance sustainable urban logistics, focusing on reverse logistics and the use of green vehicles. The research introduces a two-stage model that integrates real-time data from IoT systems to optimize vehicle routing and resource allocation across multiple echelons of logistics networks. In the first stage, an innovative model is proposed to address the production and routing of goods in urban environments, accounting for demand variability and transportation constraints. The second stage focuses on reverse logistics, where the allocation of returned goods to recovery centers is optimized to maximize resource utilization and minimize environmental impacts. The optimization models employ stochastic programming techniques, including Chance- Constrained Programming (CCP), and are solved using novel heuristic and metaheuristic algorithms. The results, validated through computational experiments and sensitivity analysis, highlight the effectiveness of these models in reducing operational costs, improving service levels, and enhancing the sustainability of urban logistics systems through the adoption of green vehicle technologies.