Genetically assisted optimization of cell layout and material flow path skeleton

A continuous plane manufacturing cell layout and intercell flow path skeleton problem formulation involving rectilinear distances between cell input/output stations is mapped to a genetic search space. Certain properties of such a search space are exploited to design a very efficient method for reduction of a mixed-integer programming problem formulation to an iterative sequence of linear programming problems. This paper reports theoretical and computational insights for efficiently finding good solutions for the above problem formulation, taking advantage of the solution structure and the search stage. The scores of the objective function on a set of test cases indicate better solutions than those previously reported in the literature. The empirical results based on multiple runs also suggest that the method generates final results that are not dependent on the quality of the initial solution; hence the solution search seems to be more global than many of the previous approaches.