Cutting path optimization in CNC cutting processes using a two-step genetic algorithm

This paper deals with the problem of generating 2D cutting paths for a stock plate nested with a set of regular and/or irregular parts. The objective of the problem is to minimize the total non-productive traveling distance of a cutter starting from a known depot, then cutting all the given parts, and returning back to the depot. A cutting path consists of the depot and piercing points, each of which is to be specified for cutting a part. The cutting path optimization problem is shown to be formulated as a generalized version of the standard traveling salesman problem. To solve the problem, a two-step genetic algorithm combining global search for piercing point optimization and local search for part sequencing is proposed. Traditional genetic operators developed for continuous optimization problems are modified to effectively deal with the continuous nature of piercing-point positions. A series of computational results are provided to illustrate the validity of the proposed algorithm.