Sequential Max-Min Ant Colony System for Traveling Salesman Problem

• Main Authors: Hsiu-Ya Chang, 張琇雅
• Other Authors: Mei-Hsiu Chi
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/76975307233622422789

碩士 === 國立中正大學 === 應用數學研究所 === 93 === Traveling salesman problem (TSP) is a typical NP-hard problem in combinatorial situation. In this paper, we introduces sequential Max-Min ant colony system (s-MMACS) which is a distributed algorithm to solve the TSP. The s-MMACS is based on two ant algorithms, ant colony system (ACS) and Max-Min ant system (MMAS). We modify the construct of ACS and use the sequential process to emphasize on the ant's running experiment. We also apply the idea of pheromone bounds to reduce the stganation. Then we can get a good approximate solution more efficiently. The results show that we can guarautee to get the appoximate solution which relative error is less than 2% in berlin52.tsp and have a high probability, 68 probability to guarantee that the solutions has a less than 6% relative error, and up to 92 probability to guarantee that the solutions has a less than 6.5% relative error. But we also show that s-MMACS has not good performance in the instances with higher density.