Solving open travelling salesman subset-tour problem through a hybrid genetic algorithm

Abstract

In open travelling salesman subset-tour problem (OTSSP), the salesman needs to traverse a set of k (≤n) out of n cities and after visiting the last city, the salesman does not necessarily return to the central depot. The goal is to minimize the overall traversal distance of covering k cities. The OTSSP model comprises two types of problems such as subset selection and permutation of the cities. Firstly, the problem of selection takes place as the salesman’s tours do not contain all the cities. On the other hand, the next problem is about to determine the optimal sequence of the cities from the selected subset of cities. To deal with this problem efficiently, a hybrid nearest neighbor technique based crossover-free Genetic algorithm (GA) with complex mutation strategies is proposed. To the best of the author’s knowledge, this is the first hybrid GA for the OTSSP. As there are no existing studies on OTSSP yet, benchmark instances are not available for OTSSP. For computational ex-periments, a set of test instances is created by using TSPLIB. The extensive computational results show that the proposed algorithm is having great potential in achieving better results for the OTSSP. Our proposed GA being the first evolutionary-based algorithm that will help as the baseline for future research on OTSSP.