الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 115 |  |  |
| | | from | 115 | | |
Abstract
Assembly lines are considered critical nodes in many production facilities that can significantly affect productivity. Therefore, researchers and practitioners have been working for decades to develop solution techniques to effectively design the assembly lines to smoothly distribute workloads among workstations without violating the assignment rules, known in the literature as the Assembly Line Balancing Problem (ALBP). The purpose of this thesis is to integrate the advantage of flexible assembly lines by considering a variety of resource alternatives for executing each task with the concept of lexicographic bottleneck assembly line balancing. The lexicographic bottleneck assembly line balancing problem guarantees a smooth workload distribution across all the workstations by minimizing the workloads of the line in descending order. In addition, considering flexible assembly lines where each workstation has a variety of equipment alternatives for performing each task increases the line flexibility to adapt with whatever customer changing demand. The proposed model could be used in designing a new assembly line or rebalancing an existing one. Accordingly, the history of the lexicographic bottleneck assembly line balancing problem (LB-ALBP) is reviewed first. A simulated annealing algorithm is developed for solving the LB-ALBP and its performance is compared to the previously developed heuristics. The results indicate that the simulated annealing algorithm gives almost the same results for most instances in less computation time. Then, the formulation of the LB-ALBP is modified by considering equipment selection decision and a new mathematical model called Lexicographic Bottleneck Mixed-Model Flexible Assembly Line Balancing Problem (LB-MFALBP) is formulated for the first time in the literature. Next, a small-sized numerical example of the LB-MFALBP is solved optimally using GUROBI solver for testing and verifying the proposed model. Then, the previously developed simulated annealing algorithm is adapted to consider the changes that have been made to the objective function and constraints. The results indicated that the proposed model gives high quality solutions in terms of the obtained results and the computational time.