Multi-objective optimization models for the design of sustainable flexible manufacturing cells

In this thesis, the topic of designing flexible cellular manufacturing systems is addressed. Along with this topic, the concept of sustainability, which includes environmental, economic, and social dimensions, is also taken into consideration. In the study, the aim is to minimize the amount of carbon emissions along with cost items including carbon emissions, inter-cell movements, machine processing, machine addition, machine removal, worker training, and additional salary (bonus). For this purpose, four different multi-objective integer mathematical programming models are developed, namely goal programming, epsilon constraint, augmented epsilon constraint (AUGMECON), and augmented epsilon constraint 2 (AUGMECON2). In these models, decisions are made regarding the selection of alternative routes for parts in each period, the assignment of workers to cells in each period, determining the number of machines assigned to, added to, and removed from cells in each period, determining the total number of workers in the system and in the cells for each period, and determining the total training time each worker will receive over all periods. The results have been obtained using the LINGO 20.0 optimization program for four different multi-objective mathematical models, utilizing the developed sample problem, and have been tested through sensitivity analyzes. Sensitivity analyzes include examining the effects of changes in values of parts demands, machine capacity, carbon emission limit, limits for the number of machines of cells, maximum number of workers in cells, and time limit that workers with skills can spend.