Optimization of Vehicle Routing Problem under Uncertainty with emphasis on Green - Lean Practices and Customer Satisfaction

Given the importance of distribution planning among the chain loops of a firm, in this research, the routing problem of cargo transport fleet is considered as the issue of transportation routing ( VRP ). Among the innovations of this model one can point, it is possible to define the penalty for violating the time window (delay and rush to delivery of the product) to ensure customer satisfaction, considering the direct relationship between weight and fuel consumption, which reducing weight, in addition to reducing costs, reduces the destructive effects of Greenhouse gas, air pollution and carbon dioxide. Reducing drivers costs, which eventually leads to lower total costs, in addition to make lean leads to Increase customer satisfaction, makes use of the concept of service time in a variety of multi product and variable conditions, taking into account the cost of product distribution .In this research, considering the Shahd-e-Pak as a case study, the distribution of product from warehouse to the consumer is modeled as a supply chain. Considering the NP-HARD problem in order to find the near optimal solutions, after setting the parameter in a comprehensive search method, the model is modeled in single-objective and multi-objective mode using Python software using heterogeneous genetic multi-objective algorithms, 2 (NSGAII) and Particle Swarm Optimization (MOPSO). At the end, these algorithms were compared with performance evaluation criteria such as runtime and the quality of the answers, and the highest algorithm was determined for each criterion. The results indicate the effectiveness of the proposed model and the superiority of using a steady-to-definitive analysis method that leads to optimization Transportation path, cost reduction and ultimately customer satisfaction. Also, by comparing the performance evaluation criteria and the duration of the implementation of the model, it was found that the MOPSO algorithm is more efficient than NSGAII.