A Decision Support System for Tobacco Distribution Partition Optimization Based on Immune Co-Evolutionary Algorithm

Abstract

The tobacco distribution in China is organized by the tobacco company in the unit of city. The distribution cycle is commonly a week with five fixed distribution districts started from the center depot, and the routes in each district are fixed for the vehicles and drivers. This method with low efficiency and high cost has continued for several decades because of the poor technology and economic reasons. In this paper, we propose a feasible and optimal method for the tobacco distribution partition balance problem (TDPBP) by breaking the fixed partitions. An immune co-evolutionary algorithm (ICEA) is proposed to search the optimal partitions. Moreover, the decision support system (DSS) for partition balance is designed. Linfen city in China as a real-world case is proved that the DSS can discover the efficient distribution planning. The comparisons among three solutions, ‘Fixed’, ‘the DSS’ and ‘Pure VRP’, further prove that the DSS is an effective decision support tool for TDPBP.