Special Research Focus on Supply Chain Linkages: Challenges for Design and Management in the 21st Century*

Supply chain management is a central and important area for academic research due to its impact on firms competing in today's global economy. Managing the flow of materials from supply sources to the ultimate customer represents a major challenge for today's managers. To assist managers, the concept of supply chain management has been adopted by many business leaders as an important way to assist in designing, planning, and controlling the network of facilities and tasks that comprise the many stages of the supply chain. In turn, the flow of academic research in the area has increased to provide a better set of guidelines for effective implementation and execution. This article sets the stage for recently completed research concentrating on supply chain management issues. First, a definition of supply chain management is provided and compared to recent usage in this area and logistics management. Also, a framework is provided that structures this dynamic and complex management task. Second, a review of past research is presented to illustrate the many paths supply chain management has traveled, and important contributions to supply management understanding and decision making. Third, recently completed research articles are introduced that have been selected to be part of this special issue of Decision Sciences. And fourth, future research directions for supply chain management that need to be pursued by interested investigators are discussed.     

Designing Distribution Systems to Support Vendor Strategies in Supply Chain Management*

The interactions among a firm's distribution strategy, market share, and distribution costs are an important consideration in the design of supply chain networks. However, these interactions are largely ignored by existing distribution system design methodologies, which assume demand is constant regardless of the firm's distribution strategy. This paper describes a multidisciplinary framework that considers these interactions in the design of “profit maximizing” distribution networks. The framework employs two major decision support methodologies: (1) binary logit models for estimating market share considering various demand-influencing parameters such as product price and distribution service, and (2) a mixed-integer programming (MIP) model for finding optimal distribution network designs. We applied the framework to an actual design problem facing a national distributor of industrial chemical products. The test results verify the framework's large-scale capability and the potential benefit of the integrated solution methodology.     

An Efficient Heuristic Procedure for Practical-Sized Capacitated Warehouse Design and Management*

The purpose of this paper is to refine and test a distribution system design model and evaluate its performance with respect to solution quality, model validation, and algorithm performance. The paper reports on the development of a multiproduct warehouse logistics model. The logistics model is used to predict the performance of distribution companies that typically have to deal with product costs, the cost to distribute the products from a set of warehouses to customer outlets, the level of service provided to its customers, and flexibility for each product group and market segment.     

Bayesian population estimation for small sample capture-recapture data using noninformative priors

One critical issue in the Bayesian approach is choosing the priors when there is not enough prior information to specify hyperparameters. Several improper noninformative priors for capture-recapture models were proposed in the literature. It is known that the Bayesian estimate can be sensitive to the choice of priors, especially when sample size is small to moderate. Yet, how to choose a noninformative prior for a given model remains a question. In this paper, as the first step, we consider the problem of estimating the population size for _Mt$M_{\mathrm{t}}$ model using noninformative priors. The _Mt$M_{\mathrm{t}}$ model has prodigious application in wildlife management, ecology, software liability, epidemiological study, census under-count, and other research areas. Four commonly used noninformative priors are considered. We find that the choice of noninformative priors depends on the number of sampling occasions only. The guidelines on the choice of noninformative priors are provided based on the simulation results. Propriety of applying improper noninformative prior is discussed. Simulation studies are developed to inspect the frequentist performance of Bayesian point and interval estimates with different noninformative priors under various population sizes, capture probabilities, and the number of sampling occasions. The simulation results show that the Bayesian approach can provide more accurate estimates of the population size than the MLE for small samples. Two real-data examples are given to illustrate the method.     

Channel Coordination for a Supply Chain with a Risk‐Neutral Manufacturer and a Loss‐Averse Retailer*

This articles considers a decentralized supply chain in which a single manufacturer is selling a perishable product to a single retailer facing uncertain demand. It differs from traditional supply chain contract models in two ways. First, while traditional supply chain models are based on risk neutrality, this article takes the viewpoint of behavioral principal–agency theory and assumes the manufacturer is risk neutral and the retailer is loss averse. Second, while gain/loss (GL) sharing is common in practice, there is a lack of analysis of GL‐sharing contracts in the supply chain contract literature. This article investigates the role of a GL‐sharing provision for mitigating the loss‐aversion effect, which drives down the retailer order quantity and total supply chain profit. We analyze contracts that include GL‐sharing‐and‐buyback (GLB) credit provisions as well as the special cases of GL contracts and buyback contracts. Our analytical and numerical results lend insight into how a manufacturer can design a contract to improve total supply chain, manufacturer, and retailer performance. In particular, we show that there exists a special class of distribution‐free GLB contracts that can coordinate the supply chain and arbitrarily allocate the expected supply chain profit between the manufacturer and retailer; in contrast with other contracts, the parameter values for contracts in this class do not depend on the probability distribution of market demand. This feature is meaningful in practice because (i) the probability distribution of demand faced by a retailer is typically unknown by the manufacturer and (ii) a manufacturer can offer the same contract to multiple noncompeting retailers that differ by demand distribution and still coordinate the supply chains.     

Strategic Outsourcing Decisions for Manufacturers that Produce Partially Substitutable Products in a Quantity‐Setting Duopoly Situation*

This article examines production and outsourcing decisions for two manufacturers that produce partially substitutable products and play a strategic game with quantity competition. When both manufacturers outsource key components to the same upstream supplier, their products become more substitutable due to the increased commonality of the products. In addition, outsourcing may create better consumer perception about the product if the manufacturers choose reputable suppliers with better brand or quality. We explicitly model the substitutability change and the brand/quality effect and provide conditions under which the manufacturers should outsource the components to a supplier. We present the subgame perfect Nash equilibriums for the situation in which there is only one supplier and the case in which two suppliers compete with each other in the upstream supply chain. Numerical examples are presented to illustrate the findings.     

Spatial Bayesian modeling of GLCM with application to malignant lesion characterization

The emerging field of cancer radiomics endeavors to characterize intrinsic patterns of tumor phenotypes and surrogate markers of response by transforming medical images into objects that yield quantifiable summary statistics to which regression and machine learning algorithms may be applied for statistical interrogation. Recent literature has identified clinicopathological association based on textural features deriving from gray-level co-occurrence matrices (GLCM) which facilitate evaluations of gray-level spatial dependence within a delineated region of interest. GLCM-derived features, however, tend to contribute highly redundant information. Moreover, when reporting selected feature sets, investigators often fail to adjust for multiplicities and commonly fail to convey the predictive power of their findings. This article presents a Bayesian probabilistic modeling framework for the GLCM as a multivariate object as well as describes its application within a cancer detection context based on computed tomography. The methodology, which circumvents processing steps and avoids evaluations of reductive and highly correlated feature sets, uses latent Gaussian Markov random field structure to characterize spatial dependencies among GLCM cells and facilitates classification via predictive probability. Correctly predicting the underlying pathology of 81% of the adrenal lesions in our case study, the proposed method outperformed current practices which achieved a maximum accuracy of only 59%. Simulations and theory are presented to further elucidate this comparison as well as ascertain the utility of applying multivariate Gaussian spatial processes to GLCM objects.     

Profust reliability of linear and circular type F and G systems having two failure criteria under Markov dependency

In this study, the profust reliabilities of (n, f, k): F(G) and < n, f, k > : F(G) systems for Markov dependent components are investigated. Having two failure criteria are the common features of these four systems. The usage of both fuzzy approach and two failure criteria in the same system provides us more realistic approach to evaluate the reliability of more complex systems. The component configurations are examined for both linear and circular sequences and the working principle of systems are studied for both F and G systems. Under all these assumptions, the profust reliabilities of (n, f, k): F(G) and < n, f, k > : F(G) systems are obtained using the distribution of run statistics. Also a new membership function different from the linear membership function which is generally used in the literature is proposed. Some numerical results which allow the comparison of the systems from various perspectives and various figures for both linear and circular type systems are presented. Some special cases (between Markov – iid assumption, conventional – profust reliability) are also considered.     

Bayesian generalized fused lasso modeling via NEG distribution

The fused lasso penalizes a loss function by the L₁ norm for both the regression coefficients and their successive differences to encourage sparsity of both. In this paper, we propose a Bayesian generalized fused lasso modeling based on a normal-exponential-gamma (NEG) prior distribution. The NEG prior is assumed into the difference of successive regression coefficients. The proposed method enables us to construct a more versatile sparse model than the ordinary fused lasso using a flexible regularization term. Simulation studies and real data analyses show that the proposed method has superior performance to the ordinary fused lasso.     

Bayes and MCMC for Undergraduates

Students of statistics should be taught the ideas and methods that are widely used in practice and that will help them understand the world of statistics. Today, this means teaching them about Bayesian methods. In this article, I present ideas on teaching an undergraduate Bayesian course that uses Markov chain Monte Carlo and that can be a second course or, for strong students, a first course in statistics.