Abstract
Although many methods have been applied to solve the combinatorial optimization problem, there are many cases in which the global optimum solution cannot be gained within a practical computation time, even if computers become more advanced. Recently,ant colony optimization (ACO) has been proposed as a meta-heuristic method to solve this problem. ACO has been tried and verified for the traveling salesman problem (TSP) to obtain the shortest path among many routes between many cities, as a representative combinatorial optimization problem. The authors have applied ACO to ship production work, such as welding and cutting, and now "marking", which is a process of writing information required in the next stage of manufacture, such as fabrication or assembly, on a steel plate by flame spray coating using a numerical control (NC) machine. In this work, the problem of searching for the optimum pathway of the NC marking torch is discussed, and a method to obtain the optimal route in a short time is presented. A system for improving ACO was developed and successfully applied to the shipyard.
