Artificial bee colony algorithm with generating neighbourhood solution for large scale coloured traveling salesman problem

Coloured travelling salesman problem (CTSP) is an extended model of multiple travelling salesman problems (MTSPs), as one kind of problem in combination optimisation problems, which has been applied to many real-world planning problems such as multi-machine engineering system (MES). Population-based algorithms such as genetic algorithm (GA) and simulated annealing (SA) algorithm are only used to solve small- or medium-scale cases in which the city number is <2000. Furthermore, in term of the convergence speed and solution quality, their performances have further improvement space. Since many real-world problems can be modelled by large-scale CTSP, it is necessary to study better algorithms to solve large-scale CTSP. Since artificial bee colony algorithm (ABC) can show good performance in solving combination optimisation problems, and therefore this study applied improved ABC algorithms to solve large-scale CTSP. The modified ABC algorithms use generating neighbourhood solution (GNS) to improve ABC for this problem, two limitation and optimisation conditions are applied into GNS during the process of reinserting cities. The extensive experiments verify that the improved algorithms can demonstrate better solution quality than the compared algorithms for large-scale CTSP.