Research on VRP Optimizing Based on Hierarchy clustering and IGA under Common Distribution

Abstract

The logistic distribution has the characteristic of dispersive customer positions, little batches and many repeated routes under common distribution. Therefore, according to the particularity of logistic distribution, the improved cluster first/route second algorithm is adopted to get solutions. Namely, the customer group can be divided into several regions using k-means algorithm in first phase. And in every region it can be decomposed into small scale subsets according with some restraint conditions using scan algorithm. In second phase, it is route optimization problems of several single TSP model. Therefore, the study proposes the improved genetic algorithm, which using individual amount control selection game in order to guarantee group diversity, using order cross operator and partial route overturn mutation operator to improve convergent speed of algorithm so as to better solve the inconsistency between diversity and convergent speed. In the end, the test proves the validity of this improved cluster first/route second algorithm combining with examples