Call number: LD2668 .T4 1985 S566 / Master of Science

In two-location inventory systems, unidirectional transshipment policies are considered when an item is not routinely stocked at a location in the system. Unlike the past research in this area which has concentrated on the simple transshipment policies of complete pooling or no pooling, the research presented in this thesis endeavors to develop an understanding of a more general class of transshipment policy. The research considers two major approaches: a decomposition approach, in which the two-location system is decomposed into a system with independent locations, and Markov decision process approach. For the decomposition approach, the transshipment policy is restricted to the class of holdout transshipment policy. The first attempt to develop a decomposition approach assumes that transshipment between the locations occurs at a constant rate in order to decompose the system into two independent locations with constant demand rates. The second attempt modifies the assumption of constant rate of transshipment to take account of local inventory levels to decompose the system into two independent locations with non-constant demand rates. In the final attempt, the assumption of constant rate of transshipment is further modified to model more closely the location providing transshipments. Again the system is decomposed into two independent locations with non-constant demand rates. For each attempt, standard techniques are applied to derive explicit expressions for the average cost rate, and an iterative solution method is developed to find an optimal holdout transshipment policy. Computational results show that these approaches can provide some insights into the performance of the original system. A semi-Markov decision model of the system is developed under the assumption of exponential lead time rather than fixed lead time. This model is later extended to the case of phase-type distribution for lead time. The semi-Markov decision process allows more general transshipment policies, but is computationally more demanding. Implicit expressions for the average cost rate are derived from the optimality equation for dynamic programming models. Computational results illustrate insights into the management of the two-location system that can be gained from this approach.

Orientador: Gladys Dorotea Cacsire Barriga / Banca: Jose de Souza Rodrigues / Banca: Moacir Godinho Filho / Resumo: Um estoque inadequado pode inviabilizar um negócio, e uma caracterização pode influenciar em seu controle e gestão. O objetivo deste trabalho é avaliar a gestão de estoque do setor de varejo calçadista através do auxílio de técnicas de Análise Multivariada e da teoria do Controle Ótimo. Foi aplicada a Análise Multivariada com a finalidade de criar índices para caracterizar as lojas de calçados; criar uma função discriminante para avaliar a diferença entre grupos de lojas com e sem excesso de estoque; utilizar esta função para classificar novas lojas nesses grupos; verificar os grupos de lojas e adaptar um modelo de controle de estoque com a Teoria do Controle Ótimo de Sethi e Thompson (2006). Para isso foram utilizados questionários para coletar dados de 30 lojas da região polo do interior do Estado de São Paulo, que foram divididas em grupos. Utilizando análise multivariada o trabalho cria dois índices de segurança através de componentes principais, para caracterizar as lojas e as variáveis mais importantes, e através da análise discriminante cria uma função discriminantes linear de Fisher para avaliar diferença significativa entre gupos e classificar as lojas, para que futuras lojas possam se alocadas aos grupos que tem excesso de estoque ou um nível adequado. Já a Teoria do Controle Òtimo foi utilizada para adaptar um modelo de controle ótimo de estoque para o setor de varejo calçadista. Foi feita uma aplicação com dados reais das 30 lojas para a análise multivariada e de uma loja para o modelo adaptado do setor de varejo calçadista do polo de Franca, e análise de sensibilidade do modelo. Com isso os objetivos foram alcançados e este estudo pode servir de apoio a estudos sobre gerenciamento de estoque e auxiliar à tomada de decisão de lojas do setor, para torná-las mais competitivas / Abstract: Inappropriate inventory may collapse a business, and a characterization can influence its control and mangement. The aim of this paper is to evaluate the inventory management of the retail footwear sector through multivariate techniques and the Optimal Control Theory. Techniques of Multivariate Analysis was applied in order to create indexes to characterize the footwear stores, create a discriminant function to evaluate the difference between groups with and without excess iventory and allocate future stores to the groups; verifying the groups of stores; and adapting a model of inventory control with Optimal Control Theory of Sethi and Thompson (2006). Questionnaires were administered to 30 stores at the region of a footwear pole on the state of São Paulo, which were divided into groups. Using multivariate analysis, the paper creates two safety indexes, through principal components, to characterize the footwear stores and the most important variable and, through discrimination analysis, creating a Fisher's linear discriminant function to evaluate a significant difference between the groups, classifying the stores, so future can be allocated to the groups with excess inventory or adequate inventory level. The Optimal Control Theory was used to adapt a model of optimal inventory control to the retail footwear sector. An application was made with actual data from 30 stores using multivariate analysis, and a store of the retail footwear sector of the region of Franca using the adapted model, and the sensitivity analysis of the model. The objectives were achieved and this study can support studies on inventory management and assist the decision-making of stores, making them more competitive / Mestre

by Stephen Wu. / Bibliography: leaves 90-92 / Thesis (M.Ph.)--Chinese University of Hong Kong, 1986

Liu, Xing. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (p. 80-87). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iv / Chapter 1 --- Introduction --- p.1 / Chapter 2 --- Literature Review --- p.10 / Chapter 3 --- The 2-Promotional Level Model --- p.15 / Chapter 3.1 --- Average Profit Function --- p.17 / Chapter 3.2 --- Auxiliary Profit Function --- p.19 / Chapter 3.3 --- Characterizing the Auxiliary Function --- p.21 / Chapter 3.4 --- Display Optimization --- p.22 / Chapter 3.5 --- Ordering Policy Optimization --- p.27 / Chapter 3.6 --- "An Algorithm for Finding an Optimal (r, Q, d, D) Policy" --- p.28 / Chapter 3.7 --- Optimality Verification --- p.30 / Chapter 4 --- Multiple Promotional Level Model --- p.36 / Chapter 4.1 --- Analysis --- p.38 / Chapter 4.2 --- "An Algorithm for Finding an Optimal (r,Q,dn, .,d1,D1,...,Dn)Policy" --- p.46 / Chapter 5 --- Extension to Random Demand Size --- p.49 / Chapter 6 --- Numerical Examples --- p.63 / Chapter 7 --- Concluding Remarks --- p.75 / Bibliography --- p.80

To Chi Kit. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (leaves 50-52). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.iii / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Structure of thesis --- p.3 / Chapter 2 --- Literature Review --- p.4 / Chapter 3 --- Problem description and formulation --- p.7 / Chapter 3.1 --- Notation --- p.9 / Chapter 3.2 --- Cost Structure --- p.11 / Chapter 3.3 --- Assumptions --- p.12 / Chapter 3.4 --- Problem Formulation --- p.14 / Chapter 4 --- Stations with deterministic demand --- p.15 / Chapter 4.1 --- Greedy Algorithm --- p.15 / Chapter 4.2 --- Example --- p.16 / Chapter 4.3 --- Properties --- p.17 / Chapter 5 --- Stations with stochastic demand --- p.21 / Chapter 5.1 --- Decision planned before arrival --- p.26 / Chapter 5.2 --- Decision made after vehicle arrival --- p.29 / Chapter 6 --- Numerical example --- p.38 / Chapter 6.1 --- Comparing decision made before and after arrival of sta- tion --- p.39 / Chapter 6.2 --- Relation between K and li --- p.40 / Chapter 6.3 --- Relation between unit penalty / cost value with K . . . --- p.40 / Chapter 7 --- Conclusion --- p.47

Warehouse operations have become an important part of the retail industry today. For many companies in the retail industry, the profit margin can be as little as one cent for every dollar sold. Because of these extremely small profit margins, it is important that companies save as much in costs along the way as they can. One such way is to cut down on warehouse costs. In this paper, we shall study the costs of storing and retrieving products in a pick, pack and ship warehouse for an office supply company. In particular, we will examine possible ways to improve the flow of products into and out of reserve racking, by implementing different storage assignment policies. One way of finding the best possible storage assignments would be to formulate a cost equation along with constraints, and to minimize the cost using linear programming techniques. We shall study the Simplex Method, and we will use it to find optimal locations for a given set of storage tasks. However, the set of products to be stored and retrieved in our warehouse changes from day to day, and is subject to customer demand. Because there is no closed form solution or algorithm to find the optimal storage assignment policy under a random situation, we will use simulation techniques to study different storage assignment policies that could be applied in the warehouse. We will study the efficiency along with the maintenance requirements for each of the different policies and compare them with the current policy being used in the warehouse today.

Chan Ying Leung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (leaves 85-89). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Importance of Reverse Auctions --- p.1 / Chapter 1.2 --- A Research Gap in Existing Literatures --- p.3 / Chapter 1.3 --- Research Focus --- p.4 / Chapter 1.4 --- Contributions of This Thesis --- p.5 / Chapter 1.5 --- Organization of This Thesis --- p.5 / Chapter 2 --- Background and Literature Review --- p.6 / Chapter 2.1 --- Review of Reverse Auctions --- p.6 / Chapter 2.1.1 --- Benefits of Reverse Auctions to Buyers --- p.9 / Chapter 2.1.2 --- Types of Reverse Auctions --- p.11 / Chapter 2.1.3 --- Implementation of the Entry-fee Reverse Auction --- p.13 / Chapter 2.2 --- Linkage between the Entry-fee Reverse Auction and the Multi- period Inventory Replenishment Model --- p.16 / Chapter 2.3 --- The Multi-period Inventory Replenishment Model with TOQC --- p.17 / Chapter 3 --- The Basic Models --- p.18 / Chapter 3.1 --- Strategic Sourcing Methodology - The Entry-fee Reverse Auc- tion --- p.18 / Chapter 3.1.1 --- Numerical Example --- p.26 / Chapter 3.2 --- Operational Procurement Methodology ´ؤ The Multi-period Inventory Replenishment Model with TOQC --- p.29 / Chapter 3.2.1 --- Numerical Example --- p.33 / Chapter 3.3 --- Chapter Summary --- p.34 / Chapter 4 --- Modifications Required for Integrating the Entry-fee Reverse Auction and the Multi-period Replenishment Model with TOQC --- p.36 / Chapter 4.1 --- Formulation of the Buyer's Expected Profit Function in the Multi-period Setting --- p.37 / Chapter 4.2 --- The Existence of Optimal TOQC --- p.38 / Chapter 4.2.1 --- Convexity of the Last-period Optimality Equation --- p.39 / Chapter 4.2.2 --- Convexity of the Two-period Problem --- p.41 / Chapter 4.2.3 --- Convexity of the N-period Problem --- p.44 / Chapter 4.3 --- The Computability of the Optimal TOQC --- p.47 / Chapter 4.4 --- Chapter Summary --- p.47 / Chapter 5 --- The Revised Model --- p.48 / Chapter 5.1 --- The Entry-fee Reverse Auction in the Multi-period Setting with TOQC --- p.48 / Chapter 5.1.1 --- Numerical Example --- p.55 / Chapter 5.2 --- Chapter Summary --- p.59 / Chapter 6 --- Numerical Analysis --- p.60 / Chapter 6.1 --- Comparison between the Fixed-quantity Reverse Auction and the Entry-fee Reverse Auction --- p.60 / Chapter 6.1.1 --- Number of Supplier --- p.63 / Chapter 6.1.2 --- Retail Price --- p.65 / Chapter 6.1.3 --- Coefficient of Variation for Demand Distribution . --- p.66 / Chapter 6.1.4 --- Average Improvement --- p.67 / Chapter 6.2 --- Duration of the Entry-fee Reverse Auction Cycle --- p.68 / Chapter 6.3 --- Chapter Summary --- p.76 / Chapter 7 --- Factors of Success for Holding the Entry-fee Reverse Auction --- p.77 / Chapter 7.1 --- Internal Organizational Infrastructure of the Buyer --- p.77 / Chapter 7.2 --- Supplier's Qualifications and Control --- p.78 / Chapter 7.3 --- Attractions of the Entry-fee Reverse Auction for Suppliers . . --- p.78 / Chapter 7.4 --- Procedural Fairness --- p.81 / Chapter 7.5 --- Total Cost Analysis --- p.81 / Chapter 7.6 --- Chapter Summary --- p.82 / Chapter 8 --- Concluding Remarks --- p.83 / Bibliography --- p.85 / Chapter A --- Order Statistics --- p.90 / Chapter B --- Conditional Order Statistics --- p.92 / Chapter C --- Virtual Marginal Cost of Procurement --- p.93

The efficiency of revenue management relies on the effectiveness of the strategies it employs. In the competing market nowadays, time has become an important concern at both demand and supply sides. With time-sensitive customers and additional benefits from intertemporal demand shift, we find that the sellers could turn to a time-differentiation based strategy as an effective revenue management tool. Motivated by real-life business issues of Toyota China dealerships, in this thesis, we consider stylized inventory-control models with delivery upgrades, in which the seller allocates its on-hand inventory to price and delivery-time sensitive customers. The seller provides two delivery-time options with different prices. Customers choose the proper purchasing option according to their heterogeneous preferences. The seller has two decisions: inventory commitment and inventory replenishment. The former addresses, within an inventory cycle, how on-hand inventories are allocated between the two classes of customers. The latter addresses, between inventory cycles, how the inventory is replenished. Furthermore, the seller may employ early delivery, namely upgrade, to achieve a higher inventory flexibility. We develop the optimal inventory allocation and upgrade, and inventory replenishment policies, and demonstrate that the optimal control can be characterized by a switching curve. Based on the basic model setting, we extend our analysis to include cases of upgrade cost, stock-out substitution, and capacity constraint. We further discuss the joint pricing and inventory decision. We obtain the form of the optimal joint policy, and show that the two strategies may well complement each other in our setting. When each is applied separately, their performances are also compared. To shed more light on the practical side, finally, we use the Toyota dealership data to calibrate the required parameters, and demonstrate the potential of the optimal inventory allocation and upgrade control. / Liang, Xiaoying. / Advisers: Houmin Yan; Youhua Chen. / Source: Dissertation Abstracts International, Volume: 72-04, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 118-124). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.

In the first model, we study the optimal procurement strategy in a two-period framework when both the spot market and the forward contract are considered. The forward contract is agreed upon in the first period, and is then delivered in the second period, when the spot market is also available. This is followed by production and demand. The objective of the buyer is to minimize his expected cost. We study the problem for two scenarios: the buyer cannot and can sell to the spot market. Through our analysis, when the buyer can not sell to the spot market, there exists a threshold forward price, under which the buyer will enter into the forward contract. This threshold is lower than the expected spot price. Furthermore, we analytically show that the optimal order quantities via forward contract increase in the mean of the spot price, but decrease in the variability of the spot price. However, the buyer only speculates using the forward contract when he can sell to spot market. / In the second model, we consider a problem in which a buyer makes procurement decisions when he faces periodic random demand and two supply sources, one is a long-term contract supplier and the other is a spot market. When he procures from the contract supplier, a fixed unit price is charged and a predetermined minimum quantity for each period must be committed, and when he procures from the spot market, a stochastic spot price plus a fixed setup cost is charged. The spot price is only realized at the beginning of each period. We show that the optimal policy consists three different (s, S) type policies. More important, we identify certain conditions under which there exist monotone properties between the policy parameters and the current spot price for a general Markov spot price process. Then, we can divide the price space into three regions, each of which corresponds to a specific policy, for each period. We also conduct numerical analysis to gain more insights into how the spot market impacts the buyer's performance. We find the buyers benefits from a more volatile market. / The last model extends the second model by incorporating an important feature that is widely seen; i.e., the procurement from the contract supplier should fulfill a total order quantity commitment (TOQC). The TOQC requires the buyer to procure no less than the predetermined commitment during the contract period, which we call the planning horizon. Thus, in each period, the buyer trades off between the possible lower cost now (by procuring from the spot market) and the reduced cost in the future (by reducing the remaining commitment). Two types of commitment contracts are considered: a minimal TOQC contract and a definite quantity contract. Our analysis characterizes an optimal procurement policy which depends on the spot price in each period and an optimal virtual remaining commitment level. Such a structured policy can be viewed as a combination of some policies of base-stock type, each of which can be computed through an equivalent system without any commitment. Moreover, some of these equivalent systems are of simple multiple-period newsvendor type. This greatly simplifies the computation of the optimal policies. We also numerically analyze how the TOQC and the spot market affects the buyer's performance. / This research develops mathematical models for inventory and procurement management in the presence of spot markets. More specifically, we consider those models by incorporating different types of supply contracts. Particular attention is paid to the quantity flexible contracts. This research is an attempt to understand how firms should adopt their operating policies in the presence of fluctuating commodity prices. In this thesis, we mainly consider the following three models. / Xue, Weili. / Adviser: Youhua Chen. / Source: Dissertation Abstracts International, Volume: 72-11, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2009. / Includes bibliographical references (leaves 122-134). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web.