Comparative analysis on optimizing performance of genetic algorithm and programming algorithm for MOTSP. Issue 6 (3rd October 2017)

• Record Type: Journal Article
• Authors: Su, Lin; Jia, Jingjing
• Is Part Of: Journal of interdisciplinary mathematics. Volume 20:Issue 6/7(2017)
• Page Start: 1533
• Page End: 1536
• Publication Date: 2017-10-03
• Subjects: traveling salesman problem -- genetic algorithm -- programming algorithm -- Matlab -- Lingo; 97P99; 16YSXK03 Applied Economics; Mathematics -- Periodicals; Mathematics; Periodicals; 510.5
• DOI: 10.1080/09720502.2017.1382202
• Languages: English
• ISSNs: 0972-0502

Abstract

Abstract: The multi-objective traveling salesman problem (MOTSP) involves one salesman to visit a set of locations (n > 1) so that each location is visited exactly once while satisfying multiple targets of distance, time, cost, etc. To solve problem like this, two different algorithms are introduced and their optimization performances are presented in details by comparisons. On one hand, the genetic algorithm (GA) implemented on Matlab was improved in this paper, which performs well in combination with rotating-disk selection with ranking, one-point crossover and mutation after similarity contrast. On the other hand, the optimization also benefits from application of 0-1 integer programming algorithm with linear and non-linear constraints where Lingo is used. Experimental results reveal both advantages and disadvantages of these two algorithms, and it is concludes that characteristics of the problem itself will decide which algorithm is more suitable when it comes to a special case.