Modelling large sequential unequal facility layout problem with improved simulated annealing solution

Abstract

The Facility Layout Problem (FLP) is an arrangement of the facilities in the floor layout. In the past research, unequal-sized FLP (UFLP) are given less attention compare to equal-sized FLP and only one type of functional tools for small problem has been explored. Thus, the research introduced a new mathematical model of Sequential Unequal Facility Layout Problem (SUFLP) that can solve large problem for different types of functional tools. In this work, the functional tools consist of machines with different function, support tools and work-in-process which are arranged sequentially in a block instead of grid structure which is commonly used in the literature. The SUFLP is solved by minimizing the total weighted sum of distances traveled from one drop-off point of a block to pick-up point of the next block and the perimeter utilization in the layout. An improved simulated annealing (SA) with the concept of tabu in the search of neighbourhood solution was applied to find the best placement of functional tools in the layout. The work proposes modified temperature setting schemes and temperature reduction rules with various initial temperatures and stopping criteria. Based on the computational experiments, the iteration based theoretically optimum using composite scheme with initial temperature of 10000 and stopping criteria of 10 consecutive non improving solutions gives the best layout design that minimizes the objective function under consideration. In order to reduce the time consumed for decision making in facility layout design, a software is in need. However, until now, only few effective and user-friendly computer aided tools have been proposed for this purpose. Moreover, the software can only cater for one type of functional tool only. Motivated by this, a software that uses SA as the solution techniques and can deal with more than one type of functional tools was developed.