Integrated application of expert systems and neural networks for machining operation sequencing

Abstract

This paper describes the development of an effective approach for solving the complex machining operation sequencing problem based on integrated application of expert system and neural network techniques. Instead of deciding the sequence for the selected operations in a pure algorithmic search manner, an expert system is used to produce partial orders that satisfy the constraints according to expert rules. A neural network is designed to determine the final operation sequence. The integrated application of the techniques enables each one of them to perform the tasks for which it is best suited. The proposed approach is executed with software simulation.