Markov kuyruk ve envanter sistemleri üzerinde duyarlılık analizi

This thesis focuses on two well-known areas in operations research: queueing and in-ventory management. The objective in both of these problems is to compensate the lossescaused by the mismatches that arise between supply (service) and demand processes. Theclass of problems we concentrate on comprises controlled queueing systems and production/inventory systems modeled by make-to-stock queues. In particular this thesis focuses onthree dynamic control policies: pricing control, admission control and stock rationing. Asthe aim of sensitivity analysis is to understand the effects of changes in system parameterson the optimal solution of a problem, we can understand how systems respond to thesechanges by sensitivity analysis. Moreover, the insight about the system obtained by theresult of sensitivity analysis can be used in the decision procedure. Since there are only afew studies on the effects of the system parameters on the optimal policies and the results ofsensitivity analysis are crucial to understand the behavior of the system when parameterschange, we aim to establish a general framework to perform sensitivity analysis on the opti-mal policies of queueing and inventory control problems. Our framework generalizes severalexisting results on specific models in the literature, and provides a structural methodologyto perform sensitivity analysis for any control problem considered here.iv