Abstract:
Assembly lines are critical for the realization of product manufacture. In recent times, there has been a shift from the make-to-stock (mass production) approach to a make-to-order (mass customization) approach and this has brought on a strong emphasis on product variety. Although variety can be included to a product at various phases of production, literature shows that by providing each functional module of the product with several variants, assembly lines provide the most cost-effective approach to achieve high product variety. However, there are certain challenges associated with using assembly lines to achieve product variety. One of these challenges is assembly line balancing. Assembly line balancing is the search for an optimum assignment of tasks, such that given precedence constraints according to pre-defined single or multi objective goal are met. These objectives include reducing the number of stations for a given cycle time or minimizing the cycle time for a given number of stations. Cycle time refers to the amount of time allotted to accomplish a certain process in an assembly process. This deviation from the optimal cycle time is technically referred to as drift. Drift can be negative or positive. Negative drift represents the time span during which an assembly line is idle, due to work being finished ahead of prescribed cycle time. Positive drift, meanwhile, represents time span in which an assembly line exceeds the prescribed cycle time. The problems caused by drift, especially positive drift, is so vast that there is a research niche are dedicated to this study called Assembly Line Balancing Problems. Various authors have proposed numerous solutions for solving assembly line balancing problems created by positive drift. However, there is very little information on optimizing multi model make-to order systems with real time inputs so as to reduce the effects of positive drift. This study looks at how such a system can be optimized by using the case study of a water bottling plant. This is done by initially looking at the literature in the field of assembly line balancing to isolate the research gap this study aims to fill. Secondly, the water bottling plant, described as the case study, is modelled using MATLAB/Simulink. Thirdly, the different optimization methodologies are discussed and applied to the created model. Finally, the optimized model is tested and the results are analysed. The results of this study show that positive drift, which can be a major challenge in a real time multi mixed assembly line, can be reduced by the optimization of assembly lines. The results of this study can also be seen as an addition to the knowledge base of the broader research on mixed model assembly line balancing.
