We propose a classification framework for the operations management literature that has looked at pricing and ordering in supply chains when supply and/or demand are uncertain. We then focus on developing three new models for managing supply and demand uncertainties in supply chains. In the first model, we study a two period sourcing problem of a firm under two sets of contracts. The contracts differ in terms of acquisition costs and the level of risk that they impose on the firm. We provide the conditions where the optimum solution is unique and also explore the behaviour of the optimum solution analytically and numerically. One application of our model is in the agribusiness supply chain and we provide numerical examples based on data from the almond industry in California. In the second model we look at a joint ordering, pricing and capacity planning problem. We characterize the optimum policy both in single and multi-period cases. In addition, we study the impact of fixed production costs on the optimum policy. The third model is devoted to coordination between a buyer and a supplier where there is a possibility of improving the supplier by both players. We analyze the problem under a Stackelberg game setting where the buyer is the leader. We show that the buyer either tries to amplify the investment of the supplier by order inflation or assumes all the investment costs. We investigate the behaviour of the optimum solution under different strategies. / Thesis / Doctor of Philosophy (PhD)

This dissertation consists of two parts with two different topics. In the first part, we investigate ``Load-Share Model" for modeling dependency among components in a multi-component system. Systems, where the components share the total applied load, are often referred to as load sharing systems. Such systems can arise in software reliability models and in multivariate failure-time models in biostatistics, for example (see Kvam and Pena (2002)). When it comes to load-share model, the most interesting component is the underlying principle that dictates how failure rates of surviving components change after some components in the system fail. This kind of principle depends mostly on the reliability application and how the components within the system interact through the reliability structure function. Until now, research involving load-share models have emphasized the characterization of system reliability under a known load-share rule. Methods for reliability analysis based on unknown load-share rules have not been fully developed. So, in the first part of this dissertation, 1) we model the dependence between system components through a load-share framework, with the load-sharing rule containing unknown parameters and 2) we derive methods for statistical inference on unknown load-share parameters based on maximum likelihood estimation. In the second half of this thesis, we extend the existing uncertain supply literature to a case where the supply uncertainty dwells in the logistics operations. Of primary interest in this study is to determine the optimal order amount for the retailer given uncertainty in the supply-chain's logistics network due to unforeseeable disruption or various types of defects (e.g., shipping damage, missing parts and misplaced products). Mixture distribution models characterize problems from solitary failures and contingent events causing network to function ineffectively. The uncertainty in the number of good products successfully reaching the distribution center and retailer poses a challenge in deciding product-order amounts. Because the commonly used ordering plan developed for maximizing expected profits does not allow retailers to address concerns about contingencies, this research proposes two improved procedures with risk-averse characteristics towards low probability and high impact events.

In this study we address a periodic review capacitated inventory system with supply uncertainty where advance demand information is available. A stochastic dynamic programming formulation is applied with the objective of minimizing the expected inventory related costs over a finite horizon. Three different supply processes are assumed. Under the all-or-nothing type supply process and partially available supply process, the structure of optimal policy is proved to be a base stock policy and numerical examples are given to demonstrate the effects of system parameters. Under Binomially distributed supply process it is shown that a simple base stock policy is not optimal.

This dissertation focuses on topics related to the value of real-time information and/or to supply uncertainties due to uncertain lead-times and yields in supply chains. The first two of these topics address issues associated with freight transportation, while the remaining two topics are concerned with inventory replenishment. We first assess the value of dynamic tour determination for the traveling salesman problem (TSP). Given a network with traffic dynamics that can be modeled as a Markov chain, we present a policy determination procedure that optimally builds a tour dynamically. We then explore the potential for expected total travel cost reduction due to dynamic tour determination, relative to two a priori tour determination procedures. Second, we consider the situation where the decision to continue or abort transporting perishable freight from an origin to a destination can be made at intermediate locations, based on real-time freight status monitoring. We model the problem as a partially observed Markov decision process (POMDP) and develop an efficient procedure for determining an optimal policy. We determine structural characteristics of an optimal policy and upper and lower bounds on the optimal reward function. Third, we analyze a periodic review inventory control problem with lost sales and random yields and present conditions that guarantee the existence of an optimal policy having a so-called staircase structure. We make use of this structure to accelerate both value iteration and policy evaluation. Lastly, we examine a model of inventory replenishment where both lead time and supply qualities are uncertain. We model this problem as an MDP and show that the weighted sum of inventory in transit and inventory at the destination is a sufficient statistic, assuming that random shrinkage can occur from the origin to the supply system or destination, shrinkage is deterministic within the supply system and from the supply system to the destination, and no shrinkage occurs once goods reach the destination.

<p>In the real world, uncertainty is a common challenging problem faced by individuals, organizations, and firms. Decision quality is highly impacted by uncertainty because decision makers lack complete information and have to leverage the loss and gain in many possible outcomes or scenarios. This study explores dynamic decision making (with known distributions) and decision learning (with unknown distributions but some samples) in not-for-profit operations and supply chain management. We first study dynamic staffing for paid workers and volunteers with uncertain supply in a nonprofit operation where the optimal policy is too complex to compute and implement. Then, we consider dynamic inventory control and pricing under both supply and demand uncertainties where unmet demand is lost leading to a challenging non-concave dynamic problem. Furthermore, we explore decision learning from limited data of focal system and available data of related but different systems by transfer learning, cross learning, and co-learning utilizing the similarities among related systems.</p>

雖然濕地是一種具有生態、經濟與景觀價值的富地資源，但由於這些價值通常具有「有行無市」的特質,因而在市場價格的論斤秤兩過程中，許多重要的濕地便成為市場價格運作下的犧牲品。為了改善此種有行無市的市場失靈現象，本研究利用可獲得使用價值與非使用價值的條件評價法(CVM)分析台北市與台南縣居民對不同濕地保育方案所賦予的價值，並利用有母數與無母數統計方法檢測此種方法可能產生的相關課題，特別是數量變異問題的檢測。本研究主要特色係納入「態度--行為」評價歷程、透過濕地意像分析掌握評價意涵與供給不確定性因素的考量，除此之外，並使用七種不同問卷結構的實驗設計檢測條件評價法的相關課題。 應用分析檢測結果顯示，台北市受訪者願付數額顯著高於台南縣，願付數額的理由不論是台北市或台南縣受訪者「非使用價值動機」皆顯著高於「使用價值動機」，而且發現保育範圍愈大的方案主觀供給機率愈小的現象。相關課題檢測結果顯示， 55公頃關渡濕地保育範圍的維護方案與8135公頃五大濕地保育範圍的維護方案受訪者最大願付數額(WTP)無顯著差異(數量變異問題)、單一濕地維護方案加上四大濕地維護方案之願付數額顯著高於五大濕地維題方案(加總問題)、「濕地信託基金」支付工具評價結果顯著高於「額外稅賦」(支付工具偏誤問題)。如果排除研究程序上的問題，此結果隱含條件評價法仍未能完全滿足數量變異的原則，就本研究實證分析探討，條件評價法未能滿足數量變異原則，供給不確定性是一項重要因素，因此條件評價法課題的改善，供給不確定效果的改善或許是一個可考量的方向。 由於條件評價法仍未能獲得可信的評價結果，而且此方法所獲得的評價結果僅是行為傾向而非賞際行為的結果，因而以條件評價法所獲得評價結果從事濕地政策的應用，在現階段仍不是一個十分恰當的做法。然而，此現象並不縮減條件評價法的重要性，因為現今仍未有比條件評價法更好的研究方法可獲得非使用價值，因此條件評價法的改善、其他評價研究方法或整合條件評價法與其他研究方法等研究發展，或許是評價研究可發展的方向。 / The major object of this study is to applyy the contingent valuation method(CVM),based on the willingness to pay(WTP)concept,to analyze the valuation of different wetland conservation projects among residents in Taipei city and Tainan prefecture. Furthermore,this study uses a set of seven surveys to test the issuess of CVM measurement of wetland conservation total values with parametric and nonparametric tests.In addition,this study incorporates the valuation process basedd on attitude-behavior studies,the images of the wetland and supply uncertainties to debate aboutthese issues． Empirical results indicate that Taipei respondents show significant higher valuation than their counterparts in Tainan,that nonuse value motives are greater than use value motives,and that the greater conservation projects have less subjective supply probabilities.Furthermore,this study finds that the magnitude variation principle does not exist in the CVM process while the aggregation problem and the payment vehicle bias exist.This study considers the supply uncertainty is one important cause of rejecting magnitude variation hypotheses．Future research should consider how to improve the supply uncertainty effect． Because the current CVM fails to achieve reliable results,and the CVM results are behavioral intention,which is different from actual behavior， thus it restricts the wetland policy pplication.However， this situation does not diminish the importance of wetland conservation valuation with the CVM,since no other methods can achieve the measurement of nonuse values． Therefore,how to improve the reliability and validity of CVM，how to integrate CVM and other methods or other valuation approaches may be one imnportant direction for future valuation research.