Newsvendor Research Articles

Production Location Selecting and Subsequent Production Decision Making for Sport Obermeyer

This paper revolved around two core research issues: production location selection and subsequent production decisions. The aim of the study was to provide a detailed analysis of Sport Obermeyer's data from 1992 to 1995 in a new way, hoping to assist companies facing similar challenges. The paper took Cost, Quality, Lead time and Minimum order quantity as bridges to the four main factors affecting the choice of production location. It detailed the advantages and disadvantages of the two production locations: Hong Kong and China. Then, with the aid of the Newsvendor model, the mismatch cost ratio for the 10 styles was determined. Ultimately, using the coefficient of variance for comparison and confirmation of the mismatch cost ratio results, the paper helped Sport Obermeyer analyze which styles of parkas had low-risk and low-uncertainty demand, and which had high-risk and high-uncertainty. The paper then combined the characteristics of the two production locations and assisted Sport Obermeyer in making subsequent production decisions. Five styles of parkas suitable for production in China and five styles suitable for production in Hong Kong were identified, ensuring stable profits and deliveries for Sport Obermeyer.

Production Location Selecting and Subsequent Production Decision Making for Sport Obermeyer

This paper revolved around two core research issues: production location selection and subsequent production decisions. The aim of the study was to provide a detailed analysis of Sport Obermeyer's data from 1992 to 1995 in a new way, hoping to assist companies facing similar challenges. The paper took Cost, Quality, Lead time and Minimum order quantity as bridges to the four main factors affecting the choice of production location. It detailed the advantages and disadvantages of the two production locations: Hong Kong and China. Then, with the aid of the Newsvendor model, the mismatch cost ratio for the 10 styles was determined. Ultimately, using the coefficient of variance for comparison and confirmation of the mismatch cost ratio results, the paper helped Sport Obermeyer analyze which styles of parkas had low-risk and low-uncertainty demand, and which had high-risk and high-uncertainty. The paper then combined the characteristics of the two production locations and assisted Sport Obermeyer in making subsequent production decisions. Five styles of parkas suitable for production in China and five styles suitable for production in Hong Kong were identified, ensuring stable profits and deliveries for Sport Obermeyer.

A Competitive Newsvendor Problem with Product Substitution under the Carbon Cap-and-Trade System

This study investigates the competitive issues of the newsvendor problem with product substitution under the carbon cap-and-trade system. Building on existing research, this paper introduces the carbon cap-and-trade system under uncertain market demand and considers that the original equipment manufacturer (OEM) can choose to procure raw materials from the contract manufacturer (CM), with both final products being substitutable. Furthermore, we explore the different substitution relationships between OEM and CM products under both pure competitive and co-opetitive modes. For this problem, decision models are developed for various scenarios, and optimal solutions satisfying given conditions are provided. We find that in one-way substitution, under pure competition, an increase in the OEM’s (or CM’s) green investment and substitution rate only leads to an increase in OEM’s (or CM’s) yields, while an increase in the OEM’s (or CM’s) green investment does not necessarily reduce CM’s (or OEM’s) yields. In the co-opetitive mode, an increase in the substitution rate and green investments of both manufacturers may lead to an increase in the yields of both manufacturers. Furthermore, an increase in carbon trading prices does not necessarily inhibit the manufacturer’s yields. Moreover, we find that under the same competition mode, under certain conditions, two-way substitution between OEM and CM can bring better profits to both manufacturers and the entire supply chain. When the two modes are in the same substitution scenario, and the CM cannot substitute for the OEM, the optimal decisions and total supply chain profits of the two modes are equal. Finally, through numerical analysis, we find that neither mode is necessarily optimal when CM can substitute for OEM. Additionally, it is observed that when the same mode is in different substitution scenarios, total supply chain profits may be enhanced in the presence of product substitution.

Fog Computing and Industry 4.0 for Newsvendor Inventory Model Using Attention Mechanism and Gated Recurrent Unit

Background: Efficient inventory management is critical for sustainability in supply chains. However, maintaining adequate inventory levels becomes challenging in the face of unpredictable demand patterns. Furthermore, the need to disseminate demand-related information throughout a company often relies on cloud services. However, this method sometimes encounters issues such as limited bandwidth and increased latency. Methods: To address these challenges, our study introduces a system that incorporates a machine learning algorithm to address inventory-related uncertainties arising from demand fluctuations. Our approach involves the use of an attention mechanism for accurate demand prediction. We combine it with the Newsvendor model to determine optimal inventory levels. The system is integrated with fog computing to facilitate the rapid dissemination of information throughout the company. Results: In experiments, we compare the proposed system with the conventional demand estimation approach based on historical data and observe that the proposed system consistently outperformed the conventional approach. Conclusions: This research introduces an inventory management system based on a novel deep learning architecture that integrates the attention mechanism with cloud computing to address the Newsvendor problem. Experiments demonstrate the better accuracy of this system in comparison to existing methods. More studies should be conducted to explore its applicability to other demand modeling scenarios.

Distributionally robust newsvendor problem with partial demand information under carbon emission regulations

Carbon emission schemes are widely adopted and implemented in practice to control greenhouse gas emissions. However, a firm restricted by carbon emission schemes with only partial demand information may face greater challenges. In this paper, we investigate an ordering and pricing newsvendor problem with partial demand information, which considers carbon emissions generated from the processes of ordering and disposing of the excess inventory. We propose three distributionally robust models to maximize the worst-case profit under different carbon emission regulations: carbon tax, cap-and-trade and carbon cap, respectively. We derive the implicit solution under the carbon tax scheme and design heuristic algorithms based on Lagrangian relaxation under the cap-and-trade and carbon cap regulations. Numerical analysis indicates: (i) all three regulations are effective in reducing carbon emissions; (ii) the implementation of carbon schemes leads to higher optimal pricing and lower optimal ordering quantity, both carbon tax and carbon cap regulations will always lead to lower profit, while the retailer may benefit from the cap-and-trade regulation by selling out the carbon credit; (iii) the proposed approach performs well under different distribution specific scenarios.

Learning Newsvendor Problems With Intertemporal Dependence and Moderate Non-stationarities

This work provides performance guarantees for solving data-driven contextual newsvendor problems when the contextual data contains intertemporal dependence and non-stationarities. While machine learning tools have observed increasing use in data-driven inventory management problems, most of the existing work assumes that the contextual data are independent and identically distributed (often referred to as i.i.d.). However, such assumptions are often violated in real operational environments where the contextual data are sequentially generated with intertemporal correlations and possible non-stationarities. By accommodating these naturally arising operational environments, our work adopts comparatively more realistic assumptions and develops out-of-sample performance bounds for learning data-driven contextual newsvendor problems.

Application of a Probabilistic Inventory Model to Industrial Band Families in the Supply Chain

Due to the high level of competitiveness that exists today, companies have the obligation to establish strategies that allow them to raise their customer service levels without incurring excessive costs. This study addresses the problems of a commercial company located in the state of Puebla, Mexico, which is in a period of restructuring, establishing strategies that allow it to reduce the losses generated from the empirical method of managing its inventories. For this purpose, the research determines the optimal lot quantity by means of the EOQ method of inventories in two of its models, eliminating the risk of shortage and comparing the results obtained in each model, being the News Boy Model with normal distribution. .

Fuzzy Multi-Item Newsvendor Problem: An Application to Inventory Management

This paper proposes a novel approach to the fuzzy newsvendor problem for inventory management applications. The main contributions of the paper are the following: a new credibility estimation is proposed, to explore the neighborhood around the most impactful demand scenarios; a simulation procedure was designed for the different demand scenarios, which allows comparison of the proposed approach with classical and fuzzy multi-item newsvendor problems; a modified genetic algorithm (GA) is introduced to ameliorate previous genetic algorithms in both the generation and evaluation of solutions. The new formulation of the fuzzy newsvendor problem, together with the modified GA, were shown to improve the average profit by up to 55% in problems with low-budget scenarios.

Adaptive, Doubly Optimal No-Regret Learning in Strongly Monotone and Exp-Concave Games with Gradient Feedback

Feasible Online Learning with Gradient Feedback Online gradient descent (OGD) is well-known to be doubly optimal under strong convexity or monotonicity assumptions: (1) in the single-agent setting, it achieves an optimal regret of [Formula: see text] for strongly convex cost functions, and (2) in the multiagent setting of strongly monotone games with each agent employing OGD we obtain last-iterate convergence of the joint action to a unique Nash equilibrium at an optimal rate of [Formula: see text]. Whereas these finite-time guarantees highlight its merits, OGD has the drawback that it requires knowing the strong convexity/monotonicity parameters. In “Adaptive, Doubly Optimal No-Regret Learning in Strongly Monotone and Exp-Concave Games with Gradient Feedback,” M. Jordan, T. Lin, and Z. Zhou design a fully adaptive OGD algorithm, AdaOGD, that does not require a priori knowledge of these parameters. In the single-agent setting, the algorithm achieves [Formula: see text] regret under strong convexity, which is optimal up to a log factor. Further, if each agent employs AdaOGD in strongly monotone games, the joint action converges in a last-iterate sense to a unique Nash equilibrium at a rate of [Formula: see text], again optimal up to log factors. The algorithms are illustrated in a learning version of the classic newsvendor problem, in which, because of lost sales, only (noisy) gradient feedback can be observed. The results immediately yield the first feasible and near-optimal algorithm for both the single-retailer and multiretailer settings.

Frontiers in Operations: Employees vs. Contractors: An Operational Perspective

Problem definition: We consider a platform’s problem of how to staff its operations given the possibilities of hiring employees and setting up a contractor marketplace. We aim to understand the operational difference between these two work arrangement models. Methodology/results: We consider a model where demand is not only stochastic but also evolving over time, which we capture via a state of the world that determines the demand distribution. In the case of employees, the platform controls the number of employee hours it uses for serving demand, whereas in the case of contractors, it sets the wage paid to them per utilized hour. We show that although the employee problem is equivalent to a standard newsvendor, the contractor one corresponds to an unusual version of the newsvendor model where utilization is the control variable. Managerial implications: This distinction makes the contractor model more flexible, allowing us to prove that it performs significantly better, especially if the order of magnitude of demand is unknown. Meanwhile, hybrid solutions that combine both employees and contractors have complex optimal solutions and offer relatively limited benefits relative to a contractor marketplace. History: This paper has been accepted in the Manufacturing & Service Operations Management Frontiers in Operations Initiative. Funding: This research was partially supported by the National Natural Science Foundation of China [Grant 71821002]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0029 .

Pricing decision in a newsvendor model with partial backorders under normal probability distribution for the demand

This paper presents a newsvendor model with backorders for customers who are willing to wait to be served. Demand follows a normal probability distribution, with the particularity that the expected value depends on the sale price and the variation coefficient is fixed. Three parameters are considered to characterize this dependence on the expected demand and the sale price: the population size of potential customers, the unit production cost of the item and an elasticity parameter with an isoelastic type. Backorders and lost sales are combined with a fixed proportion for the backorders. The quantities to be determined are the sale price and the order quantity. The goal is the maximization of the expected profit. The optimal solution is obtained in a closed form if there are no lost sales. In the case of a mixture of partial backordering and lost sales, a methodological proposal based on a numerical algorithm is given. The study reveals that the maximum expected profit and the optimal quantities to be determined are highly influenced by the unit purchasing cost and the degree of dependence of the demand concerning the sale price. Other parameters, such as the proportion of backorders and the variation coefficient of demand, are less influential. Numerical examples are used to illustrate the model, and a sensitivity analysis of the optimal solution regarding the nine initial parameters is presented. Some managerial insights deduced from the obtained results are also proposed.

Predicting mean and variance in inventory order decisions

AbstractWe develop a simple forecast‐anchoring model to explain and predict the mean and variance of observed inventory order decisions in a newsvendor problem. The model assumes that people employ a two‐step decision heuristic. In the first step, a behavioral bias may gravitate the decision maker's point forecast toward a random forecast versus a constant unbiased forecast. In the second step, a behavioral bias of the same magnitude may cause the decision maker to treat the point forecast as if it is the mean of potential demand, and then make an upward or downward adjustment depending on the underage and overage costs. We evaluate the performance of this descriptive forecast‐anchoring model across five experimental newsvendor data sets. First, we fit the model to a setting with uniform demand. We then use the corresponding estimates to generate predictions in a secondary data set with uniform demand, as an out‐of‐sample test. We proceed to fit the model to three additional newsvendor data sets, two with normal demand and one with asymmetric two‐point demand. In all cases, the model predicts the mean and variance of inventory order decisions well. We further investigate the profit implications under the forecast‐anchoring model and find that the predictions match well with the experimental data. Through improved predictions, the model can help upstream supply chain parties anticipate inventory order decisions from buyers and improve profitability.

Application of Stochastic Programming in Agricultural and Newsvendor Problems and It's Application in Real Life

The study of making the best decision under risk management in a variety of areas of our lives is known as Stochastic Programming. We will go through two-stage Stochastic Linear Programming approaches for a variety of real-world choice issues, as well as how to solve them. We will achieve this by constructing stochastic linear programming models based on real-world situations like the well-known Farmer's situation and News Vendors problems. The influence of pricing, Stochastic Integer Linear Programming problem, second stage Stochastic Integer Linear Programming problem, first stage Stochastic Binary Linear Programming problem, risk aversion problem, and continuous function for random variables based on two-stage SLP with the aid of Farmer's problem will all be examined. We will address the Newsvendor’s problem with Deterministic Equivalent Stochastic Linear Programming, an extension of Deterministic Stochastic Linear Programming for risk aversion with a high number of decision variables and restrictions, utilizing the two-stage Stochastic Linear Programming approach once more. Hand calculation is a challenging way to acquire the solution to the problems. As a result, we will use the programming language AMPL to design computer solutions for tackling both farmer and newsvendor difficulties. We will also utilize MATLAB to create graphs for the farmer's problem's continuous function. Dhaka Univ. J. Sci. 72(1): 30-45, 2024 (January)

Newsvendor model for a dyadic supply chain with push-pull strategy under shareholding and risk aversion

This paper investigates the optimization and coordination problem in the framework of classic push and pull newsvendor models under shareholding and CVaR criterion. A decentralized supply chain consisting of upstream supplier and the downstream manufacturer is considered. Firstly, taking the risk aversion into account, the optimal decision-makings are investigated under shareholding in push, pull supply chain respectively. By comparison it is found that the optimal order or production quantity depends on the degree of member’s risk aversion and shareholding ratio. Furthermore, the combined contracts are designed to coordinate the supply chain and optimal coordination parameters are obtained. Besides, the impacts of risk-averse preference and the shareholding fraction on supply chain performance are also investigated. The results indicate that in push supply chain the optimal ordering increases in the risk-averse parameter of the downstream manufacturer and the percentage of shares held by the upstream supplier in the downstream manufacturer. However, the optimal wholesale price makes the opposite change. In pull supply chain, the optimal production quantity is independent of the shareholding fraction and decreases with regards to the supplier’s risk aversion. Lastly, some numerical examples are given to illustrate the theoretical results and some suggestions to supply chain management are also provided.

A comparative analysis of how to handle stockout scenarios: the impact of newsvendor’s risk attitude

The purpose of this paper is to evaluate contrasting approaches for handling excess demand through the lens of a retailer (newsvendor) whose risk attitude (risk-neutral versus risk-averse) is modeled explicitly. We employ representative newsvendor models and provide comparison between two well-established stockout policies when demand exceeds supply. The “while supplies last” policy avoids the need for a secondary production order to satisfy the excess demand but faces potential opportunity cost through lost sales. Conversely, the “accepting backorders” policy relies on recourse production availability which is potentially costly but meets all levels of realized demand. Across distinct parameter classes, we incorporate comparison between the two policies in terms of financial metrics including expected profit and conditional value-at-risk criteria as well as metrics that relate to inventory availability and, hence, customer service, e.g., stockout probabilities and expected excess inventory. Powerful analytical results encompass both financial and inventory metrics and reveal that the outperforming policy is simply determined through relative underage cost values. That is, our insights indicate general advantages of accepting backorders when profit margins are sufficiently large and advantages of ignoring excess demand when profit margins are smaller. Although these extensive analytical takeaways hold in general, our numerical study reveals mean order quantity deviations (decision bias) in addition to which modeling approaches and counterpart optimal solutions maintain outperformance (resiliency) or lead to underperformance (sensitivity) when evaluated across all objective function criteria.

A data-driven newsvendor model for elective-emergency admission control under uncertain inpatient bed capacity.

Elective-emergency admission control referred to allocating available inpatient bed capacity between elective and emergency hospitalization demand. Existing approaches for admission control often excluded several complex factors when making decisions, such as uncertain bed capacity and unknown true probability distributions of patient arrivals and departures. We aimed to create a data-driven newsvendor framework to study the elective-emergency admission control problem to achieve bed operational efficiency and effectiveness. We developed a data-driven approach that utilized the newsvendor framework to formulate the admission control problem. We also created approximation algorithms to generate a pool of candidate admission control solutions. Past observations and relevant emergency demand and bed capacity features were modeled in a newsvendor framework. Using approximation algorithmic approaches (sample average approximation, separated estimation and optimization, linear programing-LP, and distribution-free model) allowed us to derive computationally efficient data-driven solutions with tight bounds on the expected in-sample and out-of-sample cost guaranteed. Tight generalization bounds on the expected out-of-sample cost of the feature-based model were derived with respect to the LP and quadratic programing (QP) algorithms, respectively. Results showed that the optimal feature-based model outperformed the optimal observation-based model with respect to the expected cost. In a setting where the unit overscheduled cost was higher than the unit under-scheduled cost, scheduling fewer elective patients would replace the benefit of incorporating related features in the model. The tighter the available bed capacity for elective patients, the bigger the difference of the schedule cost between the feature-based model and the observation-based model. The study provides a reference for the theoretical study on bed capacity allocation between elective and emergency patients under the condition of the unknown true probability distribution of bed capacity and emergency demand, and it also proves that the approximate optimal policy has good performance.

When Yield Is Not the Only Supply Uncertainty: Newsvendor Model of a Trade Agent

Problem definition: We study the procurement decisions of a trade agent: The agent chooses a bid (unit price to pay) to procure the goods available from seller(s). If the agent wins the bid, the supply is used to meet the buyer’s demand. Methodology/results: The trade agent’s single-period, single-product problem is a new type of newsvendor problem. We analyze the agent’s optimal bid for a seller with yield uncertainty. We show that the bid outcome distribution needs to satisfy an easy-to-check condition but no conditions on the yield distribution are needed for a unique optimal bid to exist. We also show that the expected sales-to-supply ratio that measures scarcity affects the optimal bid. We investigate the monotonicity of the optimal bid with respect to economic parameters, demand, and distributions of bid outcome and yield. The agent’s problem with multiple sellers is also a novel newsvendor network problem. For the two-seller case, we show when diversification is optimal for the agent. We show that working with both sellers may not always be optimal despite the opportunity for risk pooling and bidding only at the unreliable seller may be optimal even when the other seller is reliable. Managerial implications: We make the following recommendations for the agent: (i) bid at a seller only when the expected sales-to-supply ratio for a seller is higher than the critical ratio, considering the agent’s cost of underage and overage, (ii) increase the bid if the bid outcome distribution increases in the reversed hazard rate order, and (iii) increase or decrease the bid depending on the demand-to-supply ratio when a seller’s expected yield increases. Inclusion of additional sellers lowers the optimal bids across the seller network, but it may not be optimal to bid at all sellers. For the two-seller problem, whether to diversify is a decision easily made by computing the expected benefit of bidding at each seller. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0190 .

High-dimensional stochastic control models for newsvendor problems and deep learning resolution

AbstractThis paper studies continuous-time models for newsvendor problems with dynamic replenishment, financial hedging and Stackelberg competition. These factors are considered simultaneously and the high-dimensional stochastic control models are established. High-dimensional Hamilton-Jacobi-Bellman (HJB) equations are derived for the value functions. To circumvent the curse of dimensionality, a deep learning algorithm is proposed to solve the HJB equations. A projection is introduced in the algorithm to avoid the gradient explosion during the training phase. The deep learning algorithm is implemented for HJB equations derived from the newsvendor models with dimensions up to six. Numerical outcomes validate the algorithm’s accuracy and demonstrate that the high-dimensional stochastic control models can successfully mitigate the risk.

Managing physical inventory and return policies for omnichannel retailing

As the retail industry increasingly adopts an omnichannel strategy amidst the growth of e-commerce, this study explores the interplay between inventory and return policies in such a setting. Specifically, we focus on “return losses,” defined as the losses incurred by retailers due to customer product returns. We target how these return losses impact retailers’ profits and physical inventories. While previous research has mainly concentrated on cross-channel returns, little attention has been paid to how omnichannel return policies affect store inventory and profit. We posit that profit-maximizing retailers allow consumers to select their purchase channels based on utility but require returns to follow specific policies. In this paper, we model four return policies based on the channel of return: original purchasing channel return, offline return, online return, and cross-channel return. We use a newsvendor model to construct an optimal profit function that accounts for the additional profit from offline return, uncertainty demand, and inventory cost. Our analysis identifies the conditions under which certain return policies are beneficial or detrimental to omnichannel retailers. We discover that the retailer's product pricing, return losses, and consumer return hassle costs are the main factors influencing the best return policy and inventory policy decisions. Moreover, whether to increase or decrease store inventory and the relationship between physical inventory and return loss depend on the return policy, price, and return hassle cost. Numerical simulations support our findings, which offer practical guidance for omnichannel retailers aiming to optimize their inventory and return policies.

The value of market information and the bullwhip effect

By considering market demand information, the objective of this study is to measure the bullwhip effect in a two-stage supply chain with one supplier and multiple retailers. First, a model is detailed for measuring the bullwhip effect in which multiple retailers exhibit AR(1) demand processes and the order-up-to inventory level is calculated using market demand information. We find results that the bullwhip effect increases with the market share and lead time and decreases with the contemporaneous correlation and the number of retailers. Second, by comparing the bullwhip effect with or without market information, our model can better control and reduce the bullwhip effect. Third, we calculate the value of market information with the newsvendor model and discover that the value of market information increases with the number of retailers and the contemporaneous correlation and decreases with the market share. In addition, numerical analyses are given to verify the previous results. Moreover, some suggestions are presented to help managers make decisions on the bullwhip effect.