Layout Planning with Isles: A Genetic Approach

Abstract

Plant layout problems involve distributing different resources or departments in a given plant and achieving maximum efficiency for the services or goods being made or offered. To this end, plants are designed to optimize production flow from the first stage (i.e. as raw material) to finish product. However, optimization which is generally expressed either in terms of minimization (for example, of material handling costs) or of maximization (for example, the number of desired adjacencies in a qualitative chart) is not always feasible when real problems or real sizes are being handled. The level of complexity may turn out considerable as the number of parameters, restrictions and other variables considered in the study become larger. This kind of problem has been formulated, from a mathematical view point as a static quadratic assignment problem. However, the number of problems that are usceptible to being solved by optimization methods is very limited. Some alternatives have been called from the field of graph-theory, direct method algorithms, construction algorithms (such as CORELAP), and improvement algorithms (such as CRAFT). In this thesis work, an attempt is made to develop the algorithm for solving layout problem with real-life restriction like aisles, used in factories for the easy transfer of materials from one section to the other, using Genetic Algorithm.