Thesis Type: Postgraduate
Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Industrial Engineering, Turkey
Approval Date: 2014
Student: ÖZGÜN AKKOL
Supervisor: FATMA SEDEF MERALAbstract:
Car manufacturing is usually in the form an assembly line composed of three consecutive shops: body welding and construction, painting and finally the final assembly shop. The first stage in car manufacturing that is the body shop consists of several assembly lines in parallel each of which may have several sub-lines feeding them. Assembly line design and sequencing is therefore the heart of manufacturing a car which may have several models. Our approach mainly consists of two phases: the design phase and the sequencing phase. In the design phase of the approach, we propose an integer-programming-formulation-based robust optimization model considering the mixed-model nature of the lines. The objective of the model is the minimization of the sum of the investment and variable costs of all the assembly lines in the design and operation of the car body shop only, over the life cycle of the reference car, given the forecasted annual demand of the car and its several models, the corresponding tact time, and the available types of stations making up the lines. We obtain the optimum designs of the lines via the robust optimization model using the software GAMS, which is an extension of a single model case of the same environment. We observe lower total costs for all the lines than the total costs obtained for the single-model approach. In the second phase of our approach, we develop a genetic algorithm for the sequencing problem. The objective of mixed-model sequencing in the genetic algorithm is to have a smooth line, thus to have level utilization rates over time. The genetic algorithm provides the best fitness value for several test problems in very short computational times, when compared against the solutions obtained by the total enumeration method.