Given the transport efficiency of large vehicles in urban environments, an increasing number of enterprises are adopting the factory-warehouse-store transportation model. To address the limitations of previous models in practical transport operations, this study formulates a Two-Echelon Vehicle Routing Problem with Simultaneous Pickup and Delivery for Multi-Commodity. Furthermore, the model uniquely emphasizes certain considerations, such as prioritizing product distribution based on production batches and allowing vehicles to engage in flexible product redistribution. Considering the computational challenges arising from the substantial data involved in real-world problem instances, a hybrid heuristic algorithm is proposed. Initially, a hybrid Fuzzy C-Means is employed to decompose the problem by clustering chain stores, effectively reducing the solution space. Subsequently, an enhanced Multi-Population Genetic Algorithm, integrated with Variable Neighborhood Search, is introduced to solve the decomposed sub-problems. Experimental validation conducted with a food enterprise located in Zhengzhou, China, provides empirical support for the efficacy of the proposed model. Multiple test scenarios further illustrate the superior performance of the proposed algorithm. This study holds significant practical and theoretical implications, offering insights to aid decision-makers in reducing transportation costs and advancing the development and application of Vehicle Routing Problem models.

In smart city, the internet of vehicles provides technical support and innovative power for the development of municipal solid waste management system by virtue of the deep integration of information technology and vehicles. The vehicle routing problem based on waste classification for collection and transportation (C&T) has garnered increasing attention as a typical issue within this system. Given the diverse nature of waste types and the need to organize multiple vehicles to service collection points, congestion often arises due to simultaneous vehicle operations or queuing, posing a significant challenge. To address this, our study introduces asynchronous constraints on vehicle operations to mitigate congestion. Furthermore, building upon the separate loading and unified loading waste C&T models, a combined C&T model grounded in waste classification is devised. We propose an enhanced ant colony optimization algorithm to solve this intricate problem and validate its efficacy. Experimental findings reveal that, compared to both separate and unified schemes, the proposed combined model yields respective reductions of 5.38% and 2.75% in C&T costs. Meanwhile, the comprehensiveness of the combined model has been demonstrated through other experiments, providing a reference for the improvement and optimization of garbage classification C&T schemes.

In the process of goods delivery, it is often necessary to split and deliver (pick-up and delivery) multiple products from multiple orders. How to economically and reasonably formulate vehicle delivery routes is a challenge. Firstly, the order splitting and vehicle delivery path formulation was analyzed from the perspective of enterprises. Meanwhile, the impact of order splitting and delivery on customer satisfaction was analyzed from the perspective of consumers. Based on the above analysis, a multi-objective optimization model was established with the objectives of minimizing costs and maximizing customer satisfaction. Then, an improved NSGA-II algorithm was proposed to solve the model. In this algorithm, K-means and ant colony algorithm were used to obtain the optimal paths for two objectives, respectively. Then, the customer satisfaction problem caused by splitting was quantified using the information entropy TOPSIS method. Subsequently, the improved VNS algorithm was used to expand the solution sets of the two optimal paths for dual objective optimization. The experimental results show that the algorithm obtained Pareto front and achieved ideal experimental results.

In response to problems such as high costs and inconveniences faced by farmers in remote areas during the process of goods collection, this study proposes a two-echelon collaborative location-routing optimization transportation mode, aiming to achieve full utilization of resources and reduction of transportation costs. To solve this problem, a mathematical model is constructed with the objectives of minimizing the total cost of enterprises and maximizing the satisfaction of transfer stations. To solve this model, an algorithm integrating reachable distance-based genetic simulated annealing fuzzy C-means clustering with LOCAL SEARCH-NSGA-II is designed, hereinafter referred to as the hybrid FCM-LS-NSGA-II algorithm. The solution process consists of three steps: first, using the clustering strategy of the reachable distance-based genetic simulated annealing fuzzy C-means clustering algorithm to classify customers and initially construct distribution routes; second, optimizing the distribution routes through the local search (LOCAL SEARCH) strategy to improve the maximum satisfaction of transfer stations; finally, solving the calculation examples and optimization cases. The results show that the hybrid FCM-LS-NSGA-II algorithm has good optimization performance and can provide effective solutions for urban managers and logistics enterprises.