An Improved Path Optimization Method of Logistics Site Selection for Agricultural Products.

Tracing the processing of agricultural products from raw materials to dining tables is an important means of guaranteeing food safety. Reasonable location selection for logistics distribution centers is an effective pathway for realizing the traceability of agricultural products. However, the traditional logistics site selection for agricultural products only considers a single constraint. In order to realize the optimal path in the traceability system, the logistics site selection problem of Aksu apple was transformed into a multiobjective optimization problem. Moreover, an improved bee colony algorithm for optimal location selection with multiple constraints was proposed. A mathematical model of location selection was constructed by considering several constraints of location selection for the distribution centers and the influencing factors, such as transport distance and operation cost. The local searching ability of the solving algorithm was improved using adaptive random-search strategies. The convergence rate of this solving algorithm was increased by introducing the particle swarm optimization algorithm. The accuracy and robustness of the algorithm were improved by adjusting parameters. The optimal location of a path in the traceability system was planned by using the improved algorithm. Results demonstrate that the improved bee colony algorithm is considerably superior to the artificial bee colony (ABC) algorithm and hybrid ABC algorithm in terms of performance. It can also realize multiobjective location selection. Thus, the optimal path is planned. This study provides a good reference for the construction and resolution of the traceability location selection model in different conditional regions. [ABSTRACT FROM AUTHOR]