Order Quantity Decisions Research Articles

Can crowdsourcing improve prediction accuracy in fashion retail buying?

Fashion retailers’ buyers must decide how much to buy of merchandise well before a selling season. This order quantity decision is always challenging due to limited historical data and high demand unpredictability arising from the novelty of fashion merchandise. Despite many attempts to solve this longstanding problem, industry reports consistently show that fashion buyers’ predictions of product salability and future sales are frequently inaccurate, leading to loss of profits for retailers. In this research, the authors take an “Empirics-First” approach to explore an alternative solution, crowdsourced forecasts from ordinary customers, and investigate whether crowdsourced forecasts would be more accurate than those of expert fashion buyers and if so, how should a crowd be formed in terms of size and composition? After conducting an online experiment, finding that forecasts by a “crowd” of ordinary customers are significantly more accurate than those of expert fashion buyers, the authors test a contingency framework in a second empirical study examining how crowd size and composition impact forecasting accuracy for products of varying fashionability. The results revealed that heterogeneity in a crowd is a key factor in prediction accuracy. Specifically, crowds with more variation in income and shopping frequency made notably accurate predictions. Another key finding of the study pertains to the required crowd size; increasing the size of a crowd at first sharply decreased the crowd's prediction error. However, after a certain point, there were diminishing returns in prediction accuracy. Given the interesting results, the paper concludes with guidelines for implementing crowdsourced forecasting by fashion retailers and directions for future research.

Data-Driven Mechanisms for a Newsvendor Problem: A Case Study

Reducing food waste is paramount for a sustainable future as its implications are important to achieve sustainable development goals set by the United Nations. In many industry groups, the public awareness of reducing food waste that may potentially emerge along firms’ operations has grown. In the era of Big Data, one of the most pursued exercises of this escalating attention on reducing food waste is to utilize artificial intelligence techniques to incorporate sustainability concerns into the decision framework. Many firms embrace machine learning methods to build effective decision mechanisms that help make efficient and sustainable decisions. In this study, we analyze the impact of blending machine learning approaches with demand forecasting and order quantity decisions for a firm operating in a setting where the market demand is random, and the demand structure is not observable to the firm. The performance of the methodology is evaluated on sunflower seed demand data taken from Tadım company. Our results suggest that the joint consideration of forecasting and ordering decisions using the quantile regression approach can lead the firm to decrease its operational cost by 6% on average.

Incentive-based coordination for scheduled delivery in prefab construction

An increasing number of projects are adopting prefabrication to economize on time, labor, and materials in fixed-position layout operations, such as construction, ship building, and aircraft manufacturing. In such contexts, independent contractor and fabricator make interdependent decisions, which calls for prudent supply chain management because performance relies on coordination between their decisions. Many studies have developed integrated systems and propose various algorithms for scheduling efficiency and reliability. Nevertheless, they pay scant attention to conflicting interests amongst independent partners, which may result in subpar performance not only for the supplier but for the contractor as well. Coordination of conflicting interests has been extensively studied in economics and supply chain management; yet, those studies focus on order-quantity decisions under demand uncertainty for profit maximization, while managers in fixed-position operations are more concerned about delivery decisions under scheduling uncertainty for cost minimization. We consider the case of construction and explore a contractual scheme that aligns the agents’ decisions for coordination in a construction supply chain. Specifically, we propose a supplier rebate for coordination: the supplier grants a rebate if the contractor accepts the shipment in accordance with the delivery schedule that the contractor initially chose. We show that the optimal rebate fully coordinates the supply chain to minimize the joint supply chain costs. Thus, both the contractor and supplier benefit from the coordination by negotiating a mutually acceptable way to allocate the savings in joint costs between them. We further show that the rebate motivates the contractor to enhance its work scheduling.

Omni-channel inventory management of perishable products under transshipment and substitution

We consider a retailer selling two substitutable products through an offline and online channel. The products are stored at separate inventory locations and transshipments from the offline channel to online customers are possible, i.e. we propose a two-location inventory model incorporating customer-driven inventory interactions (demand substitution initiated by the customer) and retailer-driven inventory interactions (transshipment, i.e. demand substitution initiated by the retailer). We compare this model to a system incorporating customer-driven substitution only and use empirical data about customers’ preferences and perceptions of grocery shopping in online and offline channels to understand the trade-offs between various substitution forms in omni-channel retailing. In a numerical study, we discuss the optimal order quantities and resulting expected profits and quantify the value of a retailer-driven transshipment strategy. Although the empirical results reveal that online demands are relatively low and product substitution rates are particularly high, our results confirm that a transshipment system can achieve higher expected profits than when considering customer substitution only. Finally, we discuss the performance of policies that ignore product and/or channel substitution. We find that having complete information about product substitution is crucial for a grocery retailer’s omni-channel order quantity decision.

Manufacturer encroachment and quality investment decisions in a two‐period supply chain

AbstractWith the development of Internet technology, many manufacturers encroach into the retail market by selling products directly to consumers, which is called manufacturer encroachment. In practice, the manufacturer may encroach in normal sales period, markdown period, or both periods in a two‐period supply chain. It is a challenging task for the firm manager to provide appropriate quality level under different channel structures. This research originally studies manufacturer encroachment with endogenous quality investment decision in a two‐period supply chain considering that consumers' valuation of products depreciates over time. Using a game‐theoretic framework, this work explores four channel strategies of the manufacturer and discusses equilibrium decisions and profits. The analysis generates several insights. First, the manufacturer should promote quality investment if encroachment occurs when the direct selling is efficient. Second, regardless of which period and scenario, encroachment may increase wholesale price when the direct selling cost is smaller than a threshold. Third, results show the effect of encroachment time on price could be classified into quality investment effect, wholesale price effect, and channel‐adding effect. The quality investment effect increases retail price, while the channel‐adding effect and wholesale price effect decrease retail price. Finally, results reveal that encroachment is always the best choice for the manufacturer and it does not always harm the retailer's profits. Overall, our results and managerial insights can help the manufacturer to formulate channel strategies, wholesale price, and selling quantity decisions while help the retailer to make order quantity decision.

EXCHANGE RATE RISK IN A NEWSVENDOR FRAMEWORK UNDER A GAMMA- DISTRIBUTED EXCHANGE RATE ERROR AND ISOELASTIC DEMAND

In a global supply chain consisting of one retailer and one manufacturer, both from specific countries, when there is a time lag between the payments made whilst placing the order and the time when the order is realized, threat in the form of exchange rate fluctuation influences the optimal pricing and order quantity decisions. We explain the effect of exchange rate fluctuation underneath Gamma distribution when the retailer or producer undertakes to share the exchange rate risk and the demand error is modelled in the multiplicative form in the news vendor framework.

Application of the model combining demand forecasting and inventory decision in feature based newsvendor problem

This paper consider a one-step machine learning algorithm and propose a prediction technology using a large amount of relevant information. In the method proposed in this paper, the manager observed P features related to demand data n times, and established a prediction model through scikit-learn machine learning framework to study the data-driven newsvendor problem. In particular, the research method of this paper does not need to obtain the accurate demand distribution before optimization, and the data used by the algorithm is not limited to the historical sales data. We filter multiple characteristic variables related to the demand, substitute the filtered characteristic variable set into the model to predict the future demand, and make the order quantity decision of the newsvendor model according to the predicted demand. Finally, a case study was conduct to verify the effectiveness and feasibility of the proposed method. We find that using multiple characteristic information to predict the future demand has important guiding significance for the order quantity decision of newsvendor model, and this method can be applied to the inventory decision of many commodities.

Supply Chain Coordination under Carbon Emission Tax Regulation Considering Greening Technology Investment.

In this paper, we firstly derive the optimal strategies, including greening technology investment, production volume and order quantity decisions with stochastic demand, for the emissions-dependent supply chain composed of one manufacturer and one retailer. Then, an advance purchase discount (APD) contract and an option contract are applied to coordinate the supply chain. Moreover, an innovative prepayment-based option (PBO) contract is designed based on an APD contract and an option contract. We discuss the cash flow, the inventory risk allocation and the impacts of carbon emission tax under each contract. It is found that considering improving cash flow, preselling (or option selling) as a means of supporting the manufacturer with sufficient cash flow will help expand production and invest in greening technology. From the perspective of avoiding inventory risk, the APD contract benefits the manufacturer while the option contract benefits the retailer. However, the PBO contract generates intermediate allocations of inventory risk between manufacturer and retailer.

Inventory Management of Perishable Goods with Overconfident Retailers

In recent years, many retailers sell their products through not only offline but also online platforms. The sales of perishable goods on e-commerce platforms recorded phenomenal growth in 2020. However, some retailers are overconfident and order more products than the optimal ordering quantity, resulting in great losses due to product decay. In this paper, we apply the newsvendor model to analyze the impacts of overconfident behavior on the retailer’s optimal pricing and order quantity decisions and profit. Our model provides the overconfident retailer with a feasible and effective method to adjust optimal ordering and pricing decisions. Through numerical studies, we examine the retailer’s optimal decisions under the scenarios of complete rationality, over-estimation, and over-precision. We find that the over-estimation retailer always orders more products than the optimal order quantity, and the over-precision retailer always orders fewer products than the optimal order quantity. Under some conditions, overconfidence hurts the retailer’s revenue to a large extent. Therefore, it is beneficial for the overconfident retailer to adjust its order quantity according to our research findings.

Benefits of market information and professional advice in a vertical agricultural supply chain: the role of government provision

We study a vertical agricultural supply chain with a farmer cooperative who decides his agricultural product selling price and a social enterprise who makes her order quantity decision and resells products to the end market. The effect of market information and agricultural advice provided by the government on the system's outcomes is investigated. A two-stage game model is proposed to characterise this vertical relationship. The equilibrium decisions of the two players under different government supporting cases are derived. We contrast the system outcomes under different cases and conduct comparative analysis. It is found that (1) the farmer cooperative's profit increases if the government provides market information to him and the social enterprise at the same time; (2) the agricultural advice provided by the government can increase the benefits to each participant in the supply chain; (3) to maximise the total supply chain profit as well as its sustainability, the government may choose to provide market information to the social enterprise only; (4) when the wholesale price is intervened by the government or exogenously provided, the social enterprise's social responsibility concern can increase both the farmer cooperative's profit and the social enterprise's social payoff.

Optimal decision in MC supply chain with overconfident retailer based on the newsvendor model

In this paper, we analyze the optimal order-quantity decisions in a supply chain with mass customization (MC) manufacturer and overconfident retailers. First, we consider a newsvendor model in which an unbiased retailer sells mass customized products. The retailer needs to make order quantity decisions before the selling season. Meanwhile, the supplier is a mass customization manufacturer and implements modular production. The supply process is uncertain, as the real quantity the retailer received is the order quantity multiplied by a random yield rate. Second, two overconfident models are considered and theorems are proposed. In the first model, the behavioral bias of overconfidence only affects the retailer’s judgment of variance of market demand. In the second model, the behavior bias of overconfidence affects not only the retailer’s cognition of the variance of market demand, but also his cognition of the expectation of market demand. In addition, the relationship between the optimal decisions and the modularity level is obtained. Finally, we provide managerial insights for the decision makers of the retailers and the manufacturers on order quantity and modularity level, respectively.

Coordinating Vulnerable Supply Chains with Option Contracts

We investigate vulnerable supply chain coordination with an option contract in the presence of supply chain disruption risk caused by external and internal disturbances. The supply chain consists of a single risk-neutral supplier and a risk-averse retailer. We characterize the retailer’s order quantity decision under the Conditional Value-at-Risk (CVaR) criterion and the supplier’s production decision. The results show that facing disruption risk and risk-aversion, both the retailer and the supplier would be more prudent to order and produce less than the risk-neutral scenario, inducing damage to the supply chain performance. The number of options purchased is decreasing in disruption risk and the risk-aversion of the retailer. The supplier will increase production as the disruption risk decreases or the shortage penalty increases. When the supplier does not know the risk-aversion of the retailer, the former will produce more and bear a higher overstock risk. We also investigate conditions that facilitate vulnerable supply chain coordination and find that the existence of risk-aversion and disruption risk restrict the option price and exercise price to lower price levels. Finally, we compare the option contract with wholesale price contract from the supplier’s and retailer’s perspectives through a numerical study.

Coordination of Pricing and Order Quantity for Two Replaceable and Seasonal Products

This paper deals with the coordination of pricing and order quantity decisions for two seasonal and substitutable goods in one firm. We assume that the customers are price sensitive and they are willing to buy the cheaper products, which is known as one way and customers-based price driven substitution. First, a mathematical model is developed for one firm, which contains two replaceable products considering seasonality. The model aims to maximize the profit by determining optimal dynamic prices, order quantities and the number of periods for both of the products. Then, we show that the objective function is strictly concave of price and has a unique maximum solution. Next, an exact algorithm based on the Karush Kuhn Tucker (KKT) conditions is presented to determine the optimal decisions. Finally, a numerical example accompanied by sensitivity analysis on key parameters is developed to illustrate the efficiency of solution procedure and the algorithm.

Optimal inventory decisions for a risk-averse retailer when offering layaway

Layaway allows economically disadvantaged budget-constrained consumers to purchase expensive items through amortized payments and nominal program fees, as opposed to using high-interest financing options such as credit cards and payday loans. We consider a risk-averse retailer’s ordering decisions when offering a layaway program. We use the net loss and total loss functions, found in the literature, to determine a risk-averse retailer’s optimal order quantity under conditional value-at-risk (CVaR). We next analyze the effects of the model parameters, retailer’s risk aversion, the market default rate, enrollment fee, cancellation fee and so on, on the optimal order quantity decisions. We show that the optimal order quantity depends on different loss functions and different demand distribution functions. Further, we show that as market default rate increases or the retailer becomes more risk averse, then a rational retailer will not offer a layaway program.

A Data-Driven Functionally Robust Approach for Simultaneous Pricing and Order Quantity Decisions with Unknown Demand Function

This research considers a retailer’s problem of optimally pricing a product and making order quantity decisions without knowing the function specifying price–demand relationship. We assume that the price is set only once after collecting data, possibly from history or a market study, and that the price–demand relationship is a decreasing convex or concave function. Different from the classic approach that fits a function to the price–demand data, we propose and study a maximin framework introducing a novel concept of function robustness. This function robustness concept also provides an alternative mechanism for performing sensitivity analysis for decisions in the presence of data fitting errors. The overall profit maximization model is a non-convex optimization problem in a function space. A two-sided cutting surface algorithm is developed to solve the maximin model. An analytical approach to compute the rate of decrease of optimal profit is also given for the purposes of sensitivity analysis.

Coordination mechanism, risk sharing, and risk aversion in a five-level textile supply chain under demand and supply uncertainty

The textile supply chain has attracted worldwide attention because of its high volatility in apparel and cotton production due to coordination issue and yield uncertainty. In this context, existing analytical works related to coordination are restricted to a dyadic apparel retailer–manufacturer setting under demand uncertainty for the conventional manufacturer-led scenario. Several issues, such as the holistic depiction of the textile supply chain, impact of cotton production uncertainty, coordination mechanism for the emerging retailer-led scenario, and profitability issue of cotton firms, have not been paid enough attention. We propose an analytical model for a textile supply chain by adopting a five-level structure that comprises an apparel retailer, apparel manufacturer, textile firm, fiber firm, and cotton firm under simultaneous demand and supply uncertainty using a wholesale price contract. The wholesale price contract fails to coordinate the supply chain. Next, we show how the buyback contract and option contract coordinate this supply chain under manufacturer-led and retailer-led scenarios, respectively. Additionally, we discuss the improvement of cotton firm's profitability using a risk-sharing mechanism between the cotton firm and the fiber firm for the high loss-making scenarios. Also, we demonstrate how the risk-averse attitude of both the apparel retailers and the cotton firms can lead to the unsold cotton inventory problem. As extensions, first, we devise joint pricing and order quantity decision of apparel. Finally, we consider production uncertainties for all middle-level members in our proposed model and devise a buyback bidirectional sales rebate penalty contract for coordination.

Capacity Allocation and Compensation in a Dual-Channel Supply Chain under Uncertain Environment

The dual-channel supply chain is widely adopted by main manufacturers, potentially incurring channel conflicts between the traditional retail channel which is owned by the independent retailer and the online channel which is directly managed by the manufacturer. The purpose of this paper is to deal with the scenario where channel conflicts may arise under production capacity uncertainty, when the manufacturer tends to privilege the direct selling channel over the retail selling channel. To achieve the goal, this paper establishes a Stackelberg game model consisting of a manufacturer and a retailer, studies the scenario where the manufacturer satisfies the direct selling channel first in the presence of capacity uncertainty, employs the decision optimization and the backward induction method to find the optimal inventory decision in the direct selling channel and the optimal order quantity decision making in the retail selling channel, and designs a compensation mechanism aiming to coordinate the channel conflict in the decentralized decision-making process. Results show that the optimal decisions aiming to maximize the expected profit of each supply chain member are not able to maximize the expected profit of entire dual-channel supply chain. However, when the manufacturer compensates the retailer’s profit loss based on the unsatisfied order and, in the meantime, adjusts the wholesale price to prevent the retailer which obtains the compensation from increasing order significantly, the compensation mechanism can coordinate the decision of each supply chain member, mitigate the channel conflict, maximize the expected profit of entire dual-channel supply chain, and achieve the Pareto improvement of supply chain members’ expected profit in the decentralized decision-making process.

Who should pay for return freight in the online retailing? Retailers or consumers

This paper studies the online shopping situation where retailer faces uncertain demand and uncertain consumer valuations. We compare the suitability and effectiveness of two return freight policies, consumer affording return freight (C-Policy) and retailer paying return freight (R-Policy). Moreover, we explore the effect of non-defective returns, and the retailer’s optimal decisions on retail price and order quantity. Our results suggest that return freight policy is related to the actual quantity of returns and the proportion of non-defective returns. In general, R-Policy is reasonable when the actual returns is low, C-Policy is reasonable when the actual returns is high. But when the actual returns is not too low or too high, the return freight policy is closely related to the proportion of non-defective returns. Our study shows C-Policy is better if the proportion of non-defective returns is lower, otherwise R-Policy is better. In addition, we find the optimal retail price and the optimal order quantity decisions are also related to the actual returns and the proportion of non-defective returns. Usually, the higher the actual returns, the lower the optimal retail price and the more optimal order quantity. The higher the proportion of non-defective returns, the higher the optimal retail price and the less the optimal order quantity. At last, we find high returns are harmful to retailers, which erode the profitability of online retailers. However, an interesting observation is that the damage of high returns can be alleviated when most of returns are non-defective returns. That indicates the risk of high returns is not as terrible as we intuitively think.

Coordinating contracts under default risk control-based trade credit

This study considers a supplier–wholesaler channel in which providing trade credit to buyers incurs default risk. The motivation for this research stems from the apparel industry, where a supplier sells clothes to a wholesaler and multiple retail traders buy clothes from the wholesaler. The wholesaler provides a (down-stream) credit period to retail traders and makes optimal credit period and order quantity decisions under random and credit period dependent demand. The wholesaler assumes the risk in case they do not receive the payment for goods from a retail trader on time. In this situation, neither the wholesaler nor the supplier can receive the account owed for the same batch of clothes. Thus, the supplier provides the wholesaler a (up-stream) credit period, called default risk control-based trade credit, which decreases as the down-stream credit period increases. Based on the newsvendor model, we design coordinating contracts under default risk control-based trade credit. Our results show that trade credit policy cannot coordinate the channel, even if a buyback policy is provided simultaneously or the supplier's selling price is endogenously determined. To achieve channel coordination and Pareto optimality, two composite contracts based on partial trade credit (with buyback or quantity flexibility) are designed. Additionally, the wholesaler orders fewer product and provides a shorter credit period to retailers when the supplier considers default risk control in providing trade credit to the wholesaler. Overall, we show that the partial trade credit scheme plays an important role in channel coordination when considering default risk control-based trade credit.

Effects of Yield and Lead-Time Uncertainty on Retailer-Managed and Vendor-Managed Inventory Management

Generally, there are various elements of uncertainty in a supply chain. In particular, uncertainties in lead time, demand, and yield are very important in the semiconductor industry. Higher uncertainty can lead to bullwhip effects that can undermine the performance of the entire supply chain. This study examines the relationship between uncertainty in the supply chain and the outcome of inventory replenishment policies. Specifically, we analyze the effects of well-known uncertainties on manufacturer production quantity and retailer order quantity decisions in a decentralized supply chain. In addition, we also analyze and compare the effects of these uncertainties for the retailer-managed inventory and the vendor-managed inventory policies. Using numerical experiments, a comparative analysis of the two alternatives is conducted to determine suitable options for improving supply chain performance. In general, the performance of vendor-managed inventory is better than that of retailer-managed inventory, but we observe from the numerical experiments that there exist circumstances under which retailer-managed inventory shows better supply chain performance.