IJCSNS International Journal of Computer Science and Network Security, VOL.8 No.10, October 2008 132 Manuscript received October 5, 2008 Manuscript revised October 20, 2008 Application of MATLAB to Create Initial Solution for Tabu Search in Parallel Assembly Lines Balancing G.R.Esmaeilian, S.Sulaiman ,N.Ismail, M.M.H.M.Ahmad, M.Hamedi Department of Mechanical and Manufacturing, Faculty of Engineering, University Putra Malaysia, 43400, Serdang, Selangor, Malaysia. Abstract In comparison with the exact mathematical methods, the heuristic models are different to provide solutions close to the optimal one, saving time of processing. With the appearance of the Tabu Search by Fred Glover in 1986, different applications have been arisen from the procedure to solve different problems for the classic problems of assembly line balancing and parallel mixed model assembly line. Here, an adaptation to an existing code appears which is applied to the problem of allocating and assigning mixed model tasks for the reconfiguration of distributed generation and balance with parallel assembly line. The model provides optimal or near optimal results in terms of results obtained from calculations compared to the other methods. Keywords: Parallel assembly lines, Heuristic method, MATLAB. 1. Introduction The assembly line can be defined as the movement of the work piece from one task to the next. The tasks that must be performed on the product are divided among workers, so that each worker performs the same operation on the product, which was passed before him. The balance of an assembly line is the assignment of these tasks along a production line in order to increase the production efficiency. Many publications are available concerning the design, balancing and scheduling for Single, Multi and Mixed- Product lines. Line balancing was the main design issue in the early studies of assembly line design and is addressed in many publications [1]. The focus of this research was the single product assembly line with deterministic task times. The lines were assumed to be dedicated and were mainly balanced for a known cycle time in the simplest form [2]. The problem of balancing an assembly line is a classic Industrial Engineering problem. Even though much of the work in this area goes back to the mid-1950s and early 1960s, the basic structure of the problem is relevant to the design of production systems today, even in automated plants [3]. The assembly line balancing problem defined as assigning tasks to the workstations that minimize the amount of idle time of the line with satisfied specific condition. The first condition is that the total task time assigned to each workstation should be less than or equal to the cycle time (the time interval between two successive outputs). The second condition is the task assignments should follow the sequential processing order of the tasks [4]. Tabu search is a mathematical optimization method, fitting to the class of local search techniques that enhances the performance of a local search method by using memory structures: once a potential solution has been determined, it is marked as "tabu" (thus the name) so that the algorithm does not visit that possibility repeatedly. Tabu search is generally attributed to Fred Glover. It also has been applied successfully to maximum satisfying ability problems[5]. The performed researches in the 70s and 80s focused almost exclusively on the development of exact methods to solve the basic assembly line problem. Tabu search is a "higher level" heuristic procedure for solving optimization problems, designed to guide other methods (or their component processes) to escape the trap of local optimality. Tabu search usually has obtained optimal and near optimal solutions to a wide varieties of classical and practical problems. Tabu search (TS) is a meta-heuristic strategy for solving combinatorial optimization problems. Tabu search was introduced by Glover[5, 6] as a technique to overcome local optimality. The underlying idea is to forbid some search directions at a present iteration in order to avoid cycling, but to be able to escape from a local optimal point. This strategy can make use of any local improvement technique [7-10]. Furthermore, tabu search is an Adaptive Memory Programming (AMP) [11] that can be superimposed on many other methods. The major theme behind TS is to incorporate flexible memory (short-term and/or long-term) functions into the search procedure. Hence, the search process can avoid a move that reinstate past solutions and prevents being trapped at locally