Generic algorithm for sheet metal working tool selection, using curve matching and tree searching

Abstract

Abstract There are thousands of products made with sheet metal parts. Irrespective of the size of the part to be produced, the first step involves cutting the sheet or strip into appropriate shapes by means of the physical process of shearing. A contoured part, whether it be circular or more complex in shape, is cut between a punch and die in a press, in a process known as blanking. The same process is also used to remove internal unwanted parts of a blank, but in this case the process is known as punching. Contoured parts may also be cut by repeated small punching operations in the process known as nibbling. This paper considers punching and nibbling processes. In modern sheet metal industries the selection of a suitable set of tools for punching and/or nibbling operations, is a cost issue for two reasons. Firstly, the tooling is in itself very expensive and secondly, if the tool capacity of the tool rack is exceeded, expensive set-ups are required to mount different tools in the rack and to reprogram the system. In this work, to ‘optimise’ the process of selecting a set of tools, a simulation technique is proposed. In the simulation different sets of tools are generated by running a simulation a number of times. The set of tools with the minimum cost function value is selected. Using the well known Monte Carlo technique, the planner should be able to decide how long to run this simulation in order to reach an acceptable economic set of tools.