Ph.D. Theses
Permanent URI for this collection
Browse
Browsing Ph.D. Theses by Author "Bilgiç, Taner,"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item Competitive inventory models under demand substitution(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2012., 2012.) Fas, Genco.; Bilgiç, Taner,This work deals with the equilibrium strategies for substitutable product inventory control systems between two retailers in a nite horizon, two and single period cases. We investigate how a dynamic game framework can be used to develop various demand scenarios in a duopoly setting. We also analyze the equilibrium behavior of decentralized supply chains with competing retailers who are treated as independent agents under an e ective demand uncertainty. The agents deal with a single product, they are interested in maximizing their own pro ts, they do not share a common inventory at retail outlets and nally the excess demand is lost. There is no pricing and the competition of the game is the ordering quantities. In this thesis, we generally make nite horizon analyses including a single period result in a Stackelberg game. First, we provide a strategy applicable in a competitive environment. We show the uniqueness of the optimality while parties gave orders at the same time with a condition on the upper bound on the total number of ordering units in a nite horizon case. Furthermore, we show the existence of the optimality in a two-period model with a more general total demand which depends on two random variables. Lastly, we investigate the uniqueness of the Nash equilibirum in a single period model with a customer satisfaction measure. In addition to this, the optimality of the Stackelberg equilibrium is shown with the same customer satisfaction setting.Item Inventory policies for an assemble-to-order system with joint discount incentives(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2008., 2008.) Tombuş, Önder.; Bilgiç, Taner,We consider an assemble-to-order system to meet all of the stationary stochastic demand of a finished product in a periodic review setting. The finished product is assembled using two subassemblies (components). The demand must be met either by regular production or by using a faster but more expensive expedited mode. Components have independent setup, production, holding and expediting costs. However when both components fall short of demand they use the same expediting resource (same plane, same supplier channel, same overtime shift in a factory, etc.) causing a joint discount in unit expediting costs. This joint cost factor prevents solving of inventory control problem of each component independently and increases the time and space complexity of solving optimal inventory policy. We analyze models with and without setup costs. We prove that the optimal policy of the model without setup cost is a modified base stock policy, where target inventory for a component is a function of the other component's inventory level, both for a finite and an infinite horizon model. Similarly the optimal policy of the single and two period model with positive setup cost is a modified state dependent (s,S) policy, where (s,S) values of a component is a function of the other component's inventory level. Based on these results we develop an algorithm, which decreases time complexity, for solving finite and infinite horizon models in models without setup-costs optimally and in models with setup costs very close to optimal results.Item Joint inventory and pricing decisions: reference effects and delay sensitive customers(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2012., 2012.) Güler, Mehmet Güray.; Bilgiç, Taner,; Güllü, Refik.We consider joint inventory and pricing problem of a single product with stochastic demand in two different contexts. In the first one, we study a periodic review problem where the demand of the product is subject to reference effects. Randomness is introduced with additive and multiplicative random terms. The customers have different attitudes such as loss-aversion or loss-neutrality. For the demand models with an additive random term, we show that the problem can be decomposed into two subproblems and a steady state solution exits for the infinite horizon problem. Defining the modified revenue as revenue less production cost, we show that a state-dependent order-up-to policy is optimal for concave demand models with concave modified revenue functions. We also show that the optimal inventory level increases with the reference price. For the demand models with an additive and a multiplicative random term, we show that a state-dependent order-up-to policy is optimal for demand models and expected revenue functions which are concave after a transformation. We also provide an extensive computational study. For the second context, we consider a capacity constrained manufacturer who serves several classes of delay sensitive customers. We model the problem as an M/M/1 queueing system with non-preemptive priorities. We give closed form solutions for the inventory decisions. Using an approximation on to the problem we provide explicit solutions when there is a single customer type. We also show that the optimal prices are incentive compatible in the sense that they optimize the profit of the manufacturer even if the manufacturer does not have any information about an arriving customer and let the customer choose a price from a provided menu of prices.Item Maintenance of a multi-component dynamic system under partial observations(Thesis (Ph.D.)-Bogazici University. Institute for Graduate Studies in Science and Engineering, 2008., 2008.) Ünlüakın, Demet Özgür.; Bilgiç, Taner,This thesis studies the maintenance of a dynamic system consisting of several components which age in time at a given failure rate. The states of the components are hidden. In each decision epoch, the decision of whether replacing a component or doing nothing is to be made. The major difference of this problem from the other main- tenance problems is its complex structure due to many components. Two versions of the maintenance problem are studied. In the first one, it is possible to estimate the re- liability of the whole system. The aim is to find a minimal maintenance cost given that the reliability of the system should always be above a predetermined threshold value. In the second problem, partial observations, i.e., signals related with the components are observed in each time period. The next observation may have an associated cost to the decision maker. This problem is a partially observed Markov decision process (POMDP). Dynamic Bayesian networks (DBNs) are proposed as a solution to the first prob- lem. Four heuristic approaches are presented to select the component to be replaced. A hierarchical heuristic solution procedure is proposed to solve the second problem. An aggregate model is developed by aggregating states and actions so that it can be solved with exact POMDP solvers. Disaggregation is done by simulating the process with a DBN and applying troubleshooting approaches in the decision epochs where replacement is planned in the aggregate policy.