Study on Urban Logistics and Distribution Path and Carbon Emission Balance Based on Multi-objective Optimization

Abstract

With the growing demand of urban logistics and distribution, how to optimize the distribution path to minimize the cost and carbon emission has become an important issue. In this paper, a multi-objective optimization model is proposed for an urban logistics and distribution problem, aiming to balance the distribution cost and carbon emission. Specifically, this paper firstly proposes a static path optimization model to design the optimal distribution routes under fixed traffic conditions using the mixed integer linear programming (MILP) method. Second, Genetic Algorithm (GA) and Dynamic Programming (DP) are used for real-time distribution paths when dynamic traffic changes (e.g., changes in traffic flow during morning and evening peaks) are considered. Finally, for the trade-off relationship between distribution cost and carbon emission, this paper constructs a multi-objective optimization model and adopts the weighted sum method to balance these two objectives. By setting appropriate weight coefficients, this paper obtains multiple Pareto-optimal solutions and provides path selection schemes under different weight configurations. In the solution process, genetic algorithm and particle swarm optimization (PSO) are used to find the optimal distribution paths and combined with constrained optimization techniques to ensure that the path selection traffic rules and other constraints. The experimental results in this paper show that reasonable path selection can significantly carbon emissions while optimizing the distribution path. In addition, the dynamic traffic strategy considered in the model can also effectively deal with the traffic changes during peak hours. The research in this paper provides valuable theoretical support for the optimization of urban logistics and distribution paths, and provides concrete implementation solutions for practical applications such as balancing cost and environmental protection goals. The results show the trade-offs between different paths, which can provide decision support for logistics companies in multi-objective optimization and have strong practical application value.