Tool selection for sheet metal punching operations - searching techniques

摘要

In recent years it has become widely acknowledged that integrating the wide spectrum of product design and manufacturing issues brings about an improvement in product development time and cost, as well as helping to achieve higher product quality. During this research an interest was taken in developing a design-supporting tool which could be used at the design level as well as at the planning level. As the reader will appreciate there are many thousands of products made from sheet metal. However, despite the variation in the range of the shapes and sizes of parts produced by sheet metal processes it is usually found that almost any shape can be produced by use of one or more common processes such as shearing, punching, nibbling, laser, drawing or bending. This paper discusses the problem of selecting the best set of tools to punch or nibble a given 2D-sheet metal part. In this work, it was found that the problem in hand is a problem of optimising an explosive decision tree [1-4]. Any attempt to generate an analytical solution by mathematically formulating the problem would probably require the imposition of some restrictions on definition of the problem, which would then result in a tailored solution rather than a generic one. Thus, a new approach to tool selection was taken in which the Monte Carlo method was investigated. The approach showed very promising results indicating the feasibility of such an approach.