Robotic assembly line design with tool changes

This thesis is focused on assembly line design problems in robotic cells. The mixed-model assembly line design problem that we study has several subproblems such as allocating operations to the stations in the robotic cell and satisfying the demand and cycle time within a desired interval for each model to be produced. We also ensure that assignability, precedence and tool life constraints are met. The existing studies in the literature overlook the limited lives of tools that are used for production in the assembly lines. Furthermore, the studies in the literature do not consider the unavailability periods of the assembly lines and assume that assembly lines work 24 hours a day continuously. In this study, we consider limited lives for the tools and hence we handle tool change decisions. In order to reflect a more realistic production environment, we deal with designing a mixed-model assembly line that works 24 hours a day in three 8-hour shifts and we consider lunch and tea breaks that are present in each shift. This study is the first one to propose using such breaks as tool change periods and hence eliminate tool change related line stoppages. In this setting, we determine the number of stations, operation allocations and tool change decisions jointly. We provide a heuristic algorithm for our problem and test the performances of our heuristic algorithm and DICOPT and CPLEX solvers included in GAMS software on different instances with varying problem parameters.