Traffic network distribution based on distribution center problem and genetic algorithm

• Published in: 2012 IEEE International Conference on Information Science and Technology pp. 219 - 223
• Main Authors: Wei Xu, Ren-Jie Shen, Gui-Fang Wu, Kang Zhou
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.03.2012
• Subjects: Biological cells; Educational institutions; Encoding; Genetic algorithms; Layout; Logistics; Time factors
• ISBN: 9781457703430; 1457703432
• ISSN: 2164-4357
• DOI: 10.1109/ICIST.2012.6221641

The first traffic network distribution based on distribution center problem (TNDBDCP) is put forward, which can not be solved by traditional algorithms. In order to solve TNDBDCP, improved genetic algorithm is put forward based on the idea of global and feasible searching. In the improved genetic algorithm, chromosome is generated to use binary-encoding, and more reasonable fitness function of improved genetic algorithm is designed according to the characteristics of spanning tree and its cotree; in order to ensure the feasibility of chromosome, more succinct check function is introduced to three kinds of genetic operations of improved genetic algorithm (generation of initial population, parental crossover operation and mutation operation); three kinds of methods are used to expand searching scope of algorithm and to ensure optimality of solution, which are as follows: the strategy of preserving superior individuals is adopted, mutation operation is improved in order to enhance the randomness of the operation, crossover rate and mutation rate are further optimized. The validity and correctness of improved genetic algorithm solving MSTLCP are explained by a simulate experiment where improved genetic algorithm is implemented using C programming language. And experimental results are analyzed: selection of population size and iteration times determines the efficiency and precision of the simulate experiment.