Solving a fixture configuration design problem using genetic algorithm with learning automata approach

Abstract

Proper fixture design is crucial to workpiece quality assurance in manufacturing. Incorrect fixture design may lead to workpiece deformation during machining. The fixture configuration design is one of the important aspects of fixture design. This paper deals with fixture layout optimization problem. The objective is to minimize the norm of all the passive contact forces satisfying Coulomb friction constraint, work-piece static equilibrium constraint and contact constraint, for the entire cutting operation. To solve this problem, the paper proposes Genetic Algorithm with Learning Automata (GALA) algorithm, which is a population based interconnected learning automata algorithm incorporating genetic operators. The algorithm enjoys the good characteristics of both GA and LA. It is validated with an example of face milling operation. The optimal layout is found to be in tune with empirical facts. Also, for the further investigation of the algorithm, it has been tested on a different problem sets and a comparative study is carried out.