Retailer Orders Research Articles

The evolution of two‐stage production and order equilibrium in a random yield supply chain with demand information updating

AbstractThe mismatch between production, order, and demand seriously affects supply chain performance. However, most research focus on the mismatch between the retailer's order and customer's demand, which ignores the influence of the supplier's random yield on supply chain members' decision‐making. This paper investigates a two‐stage optimization problem within a two‐echelon supply chain, featuring a supplier with random yield and a retailer updating demand information in real‐time. Faced with a long production lead time, the retailer can either place advance orders at the production season's onset (first‐stage advance order) or opt for instant orders at the beginning of the sales season (second‐stage instant order). To ensure timely order fulfillment, the supplier initially employs a cost‐effective regular production mode with random yield during the production season. If yields are insufficient during sales, a pricier emergency production mode with guaranteed output becomes available. Utilizing a dynamic programming approach, we formulate the two‐stage optimization problem to derive optimal production and order decisions. Our analysis uncovers how realized random yield and stochastic market signals influence emergency production and instant order quantities in the second stage. We compare expected profits in scenarios with perfect and imperfect market signals, probing the members' preferences regarding order strategies. An intriguing finding emerges: as instant wholesale prices rising, the supplier's preferred order strategy diverges from the retailer. By strategic adjustments to the instant wholesale price, we demonstrate the potential for unanimous agreement on preferred order strategies among supply chain members — a quality enhancing the chain's flexibility and performance. Moreover, we extend the model to hybrid order strategies and identify conditions for unanimous preference among the three strategies. To bolster our theoretical findings, we provide numerical examples, lending practical support to our study.

The information content of retail order flow: Evidence from fragmented markets

We examine the information content of off-exchange retail order flow relative to off-exchange institutional and on-exchange order flow. We use alternative sources of retail order identification and account for the fact that market opacity affects the order routing choices of both institutions and the brokers who sell retail orders. After controlling for volume effects, we find that retail order flow is more informed than off-exchange institutional order flow. Off-exchange price discovery comes mostly from retail order flow. However, on days with greater Robinhood activity, the information content of retail order flow drops. Our results reconcile some mixed findings in this literature.

Optimal Policies and Heuristics to Match Supply with Demand for Online Retailing

Problem definition: We consider an online retailer selling multiple products to different zones over a finite horizon with multiple periods. At the start of the horizon, the retailer orders the products from a single supplier and stores them at multiple warehouses. The retailer determines the products’ order quantities and their storage quantities at each warehouse subject to its capacity constraint. At the end of each period, after random demands in the period are realized, the retailer chooses the retrieval quantities from each warehouse to fulfill the demands of each zone. The objective is to maximize the retailer’s expected profit over the finite horizon. Methodology/results: For the single-zone case, we show that the multiperiod problem is equivalent to a single-period problem and the optimal retrieval decisions follow a greedy policy that retrieves products from the lowest-cost warehouse. We design a nongreedy algorithm to find the optimal storage policy, which preserves a nested property: Among all nonempty warehouses, a smaller-index warehouse contains all the products stored in a larger-index warehouse. We also analytically characterize the optimal ordering policy. The multizone case is unfortunately intractable analytically, and we propose an efficient heuristic to solve it, which involves a nontrivial hybrid of three approximations. This hybrid heuristic outperforms two conventional benchmarks by up to 22.5% and 3.5% in our numerical experiments with various horizon lengths, fulfillment frequencies, warehouse capacities, demand variations, and demand correlations. Managerial implications: A case study based on data from a major fashion online retailer in Asia confirms the superiority of the hybrid heuristic. With delicate optimization, the heuristic improves the average profit by up to 16% compared with a dedicated policy adopted by the retailer. The hybrid heuristic continues to outperform the benchmarks for larger networks with various structures. Funding: X. Li is supported by the Singapore Ministry of Education [Tier 1 Grant 23-0619-P0001]. Y. F. Lim is grateful for the support from the Singapore Management University under the Maritime and Port Authority Research Fellowship and the Singapore Ministry of Education [Tier 1 Grant MSS23B001]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2021.0394 .

Allocating Inventory Risk in Retail Supply Chains: Risk Aversion, Information Asymmetry, and Outside Opportunity

Problem definition: Recent global crises have caused unprecedented economic uncertainty and intensified retailers’ concerns over inventory risks. Mitigating inventory risks and incentivizing retailer orders is critical to managing retail supply chains and restoring their norms after severe impacts. We study the allocation of inventory risk using contracts in a retail supply chain with a risk-neutral manufacturer and a risk-averse retailer. We consider two factors that affect the effectiveness of contracting: (1) asymmetric risk aversion information—retailers’ attitudes are typically diverse and unknown to the manufacturer, and (2) uncertain outside opportunity—retailers typically face a volatile external business environment. Methodology/results: With a game-theoretic model that captures the interaction among risk aversion, information asymmetry, and outside opportunity, we derive the contracting equilibrium under two widely adopted risk allocation schemes—push (i.e., the retailer bears the inventory risk) and pull (i.e., the manufacturer bears the inventory risk) contracts. Contrary to the conventional wisdom that pull contracts are more effective in risk mitigation, we show that push contracts may induce larger expected order quantities and achieve the highest supply chain efficiency due to the interaction of asymmetric risk aversion information and risky outside opportunities. We also find that the manufacturer may obtain higher profits with push contracts when both the heterogeneity in the retailer’s risk attitude and the risk of the outside opportunity are sufficiently high. In addition, when the risk of the outside opportunity is in a medium range, the push contract allows the manufacturer to fully eliminate the information rent and achieve the supply chain’s first-best outcomes. We further evaluate the effects of product profitability and demand uncertainty and generalize the retailer’s risk measure to any coherent risk measure. Managerial implications: Our analysis highlights the importance of modeling asymmetric risk aversion information and risky outside opportunities in analyzing supply chain contracting. When considering these practical factors, allocating more inventory risks to a risk-averse retailer may be better than a risk-neutral manufacturer. Our results provide novel insights into the selection of proper contract types for managing inventory risks in retail supply chains. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2022.0624 .

Feedback Trading: The Intraday Case of Retail Derivatives

ABSTRACTWe analyze retail order flow in terms of intraday feedback trading patterns. Using a unique data set of exchange trades and high‐frequency quotes, we first provide evidence that retail investors actively and consciously respond to short‐term intraday returns in a negative feedback, contrarian fashion. Second, we show that some retail investors also feedback trade on tick‐by‐tick returns. Third, we find that on average this behavior leads to significant losses on the day they open a position. These losses are primarily due to the bid‐ask spread and to investors' timing inability, but not to market makers taking advantage of investors.

Revenue sharing contract in competitive supply chain network coordination: location based on VMI policy and bullwhip effect

Online shopping advances have made the timely delivery of goods to the place of the buyer a necessity. This study considers a significant and rival supplier with a retailer network, and we use Stackelberg game to solve the problem. The major supplier (the leader) is connected to retailers and covers their orders through a Vendor Management Inventory policy and revenue-sharing contract. The major supplier pays fines in case of any shortage, and minimises the loss of retailers in the network through good warehouse locations. The rival supplier (the follower) signs contract with some retailers encouraging them to receive their orders to achieve higher utility. Hence, the expected utility function of both suppliers is calculated, and their strategies are measured at the equilibrium point in a zero-one model. Furthermore, a better view of major decisions by using bullwhip effect (BWE) analysis in the competition between main and rival suppliers is done. Results indicate a lower BWE when the actual order of the retailer equals the retailer's order from the supplier's perspective, and the demand variance ratio is less than the contracts’ profit variance ratio. Moreover, it is possible to control the demands variance ratio by controlling the ratio of contracts’ profit.

Exact evaluation and optimization of integer-ratio policies for stochastic one-warehouse multiple-retailer inventory systems

This paper considers one-warehouse multiple-retailer inventory systems. The retailers are non-identical and face Poisson demand processes. We consider an integer-ratio policy in which each retailer orders in a fixed interval that is either an integer or integer-ratio multiple of the warehouse's reorder interval. Base-stock policy and virtual allocation are assumed. We show that cost evaluations at the retailers are completely separable, leading to a simple approach to develop an optimal integer-ratio policy for a two-level distribution system with heterogeneous retailers.

The Variance Learning Curve in Retail Order Picking

The notion that prior experience can improve future performance has reached general acceptance in the operations and supply chain management literature. However, prior experience may not only have a positive impact on average performance but in addition also on the variance of performance. We empirically examine this issue by considering the performance of 19 temporary workers in retail order picking new to this task, performing 22,603 storage location visits and use a two-stage estimation method of a heteroskedastic learning curve model. In the first stage, we estimate the impact of prior experience on task performance time. For the second stage, we extract the error terms from the first stage model and use it as the dependent variable in the second stage. Through this, we can estimate the impact of prior experience on performance variability. Our first stage model suggests that prior experience can improve average task performance on a system level for all new temporary order pickers observed. However, on an individual level, these learning curves are heterogeneous, with varying intercepts, slopes, and functional forms. Going beyond the impact of experience on average performance, our second stage model finds that one week of experience working in the examined warehouse can reduce performance variability by 26%, but at diminishing increments. These findings suggest that prior experience can mitigate variability in task performance time, which indicates that the recently discussed variance learning curve exists for order picking tasks. Our results allow managers to make informed decisions on how to benefit from the positive effects of prior experience on average performance and performance variability when employing a temporary workforce for the example case of retail warehouse order picking.

Impact of Supply Risk Sharing on Supply Chains: Multiplicative vs. Additive Risks

Abstract We consider a supply chain in which a retailer orders from a manufacturer(s) who face(s) a stochastic supply risk (random yield) under single or dual-sourcing cases. In specific, we look into this problem in two different yield risks: multiplicative and additive. One might intuit that if the retailer shares a manufacturer’s random yield risk with the manufacturer, the manufacturer will be better off. Interestingly, we confirm that this intuition is only valid in the additive yield risk but not necessarily in the multiplicative yield risk. Moreover, under dual sourcing, both manufacturers fiercely compete on their prices (i.e., Bertrand-like competition) to become the sole source in the additive yield risk, but the manufacturers do not compete as much in the multiplicative yield risk. Lastly, this paper shows that the supply chain inefficiency may decrease (increase) as risk-sharing increases in the additive risk model under dual sourcing (single sourcing) while it does not changeinthemultiplicativeriskmodel. Key words: supply risk, random yield, dual sourcing, game theory JEL Classification: L11, M11, M21

Deep Reinforcement Learning for One-Warehouse Multi-Retailer inventory management

The One-Warehouse Multi-Retailer (OWMR) system is the prototypical distribution and inventory system. Many OWMR variants exist, e.g. demand in excess of supply may be completely back-ordered, partially back-ordered, or lost. Prior research has focused on the study of heuristic reordering policies such as echelon base-stock levels coupled with heuristic allocation policies. Constructing well-performing policies is time-consuming and must be redone for every problem variant. By contrast, Deep Reinforcement Learning (DRL) is a general purpose technique for sequential decision making that has yielded good results for various challenging inventory systems. However, applying DRL to OWMR problems is nontrivial, since allocation involves setting a quantity for each retailer: The number of possible allocations grows exponentially in the number of retailers. Since each action is typically associated with a neural network output node, this renders standard DRL techniques intractable. Our proposed DRL algorithm instead inferences a multi-discrete action distribution which has output nodes that grow linearly in the number of retailers. Moreover, when total retailer orders exceed the available warehouse inventory, we propose a random rationing policy that substantially improves the ability of standard DRL algorithms to train good policies because it promotes the learning of feasible retailer order quantities. The resulting algorithm outperforms general-purpose benchmark policies by ∼1−3% for the lost sales case and by ∼12−20% for the partial back-ordering case. For complete back-ordering, the algorithm cannot consistently outperform the benchmark.

When is the next order? Nowcasting channel inventories with Point-of-Sales data to predict the timing of retail orders

Slow-moving goods are common in many retail settings and occupy a vast part of retail shelves. Since stores sell these products irregularly and in small quantities, the replenishing distribution center may only place batched orders with manufacturers every few weeks. While order quantities are often fixed, the challenge for manufacturers facing such intermittent demand is to forecast the order timing. In this paper, we explore the value of Point-of-Sales (PoS) data to improve a food manufacturer’s order timing forecast for slow-moving goods. We propose an inventory modeling approach that uses the last order, PoS data from retail stores, and the expected lead time demand to estimate the retailer’s channel inventory. With this dynamic estimate, we can ‘nowcast’ the retailer’s inventory and predict his next order. To illustrate our methodology, we first conduct an experimental simulation and compare our results to a Croston variant and a moving average model. Next, we validate our approach with empirical data from a small German food manufacturer that serves a grocery retailer with a central distribution center and 53 hypermarkets. We find that, on average, our approach improves the accuracy of order-timing predictions by 10–20 percent points. We overcome a shrinkage-induced bias by incorporating an inventory correction factor. Our approach describes a new way of utilizing PoS data in multi-layered distribution networks and can complement established forecasting methods such as Croston. Particular applications arise when the order history is short (e.g., product launch) or represents a bad predictor for future demand (e.g., during COVID-19).

Retail Trading in Options and the Rise of the Big Three Wholesalers

ABSTRACTWe document a rapid increase in retail trading in options in the United States. Facilitated by payment for order flow (PFOF) from wholesalers executing retail orders, retail trading recently reached over 60% of total market volume. Nearly 90% of PFOF comes from three wholesalers. Exploiting new flags in transaction‐level data, we isolate wholesaler trades and build a novel measure of retail options trading. Our measure comoves with equity‐based retail activity proxies and drops significantly during U.S. brokerage platform outages and trading restrictions. Retail investors prefer cheaper, weekly options with average bid‐ask spread of 12.6%, and lose money on average.

Optimizing retailer’s order and financing decisions on an e-commercial platform considering cash flow

The mismatch in cash flow often distorts business operational decisions and even lead to bankruptcy for enterprises. This paper investigates the order and financing decisions of a capital constrained retailer who borrows from an e-commerce platform to fund its business operations. The ecommercial platform, which has full capital, provides three financing schemes: (1) lump-sum repayment (scheme L), (2) average capital plus interest repayment (scheme P), and (3) average capital repayment (scheme A). We first model the financing behaviors of the retailer and determine the circumstances in which the retailer favors a specific financing scheme. Then, we propose a viable cash flow matching strategy in which the retailer retains a portion of its initial capital to address potential repayment shortfalls in each period. To the best of our knowledge, this paper is the first to integrate the capital constrained retailer’s cash flow management into the platform financing scheme. The results show that: In the absence of cash flow considerations, the retailer prefers scheme L since selecting scheme P or A may lead to bankruptcy. In contrast, if the cash flow matching is efficiently realized, the retailer always prefers scheme P to enhance its performance. Numerical examples are used to validate the theoretical results.

Inventory management for stockout-based substitutable products under centralised and competitive settings

Inventory planning in fashion markets is highly challenging, owing to uncertain demand; yet, in making inventory decisions, retailers may be able to capitalise on high substitutability between products. This research develops single-period inventory-management models describing a market with two substitutable products, under stockout-based substitution; i.e. when a customer’s preferred product is out-of-stock, s/he may choose to purchase the substitute. Two settings are considered: centralised (a single retailer who sells both products) and competitive (two retailers, each selling one product). For each setting, we derive closed-form analytical solutions for the inventory levels that maximise expected profit. The model is further enriched with sales data from an online apparel retailer offering substitutable products (a sneaker in different colours), and we analyse the sensitivity of the optimal inventory levels and profits to parameter values. Key findings include the following: (i) Under competitive conditions, both retailers always order positive inventory so as not to lose customers. However, in a single-retailer setting, there are situations in which the retailer orders inventory for only one product. (ii) The optimal inventory levels and corresponding profits are highly sensitive to consumers’ willingness to substitute between products. These findings provide concrete insights that can guide fashion brands’ inventory-management decisions.

Resolving a Paradox: Retail Trades Positively Predict Returns but Are Not Profitable

Abstract Retail order imbalance positively predicts returns, but on average retail investor trades lose money. Why? Order imbalance tests equal-weighted stocks, but retail purchases concentrate on attention-grabbing stocks that subsequently underperform. Long–short strategies based on extreme quintiles of retail order imbalance earn dismal annualized returns of −14.8% among stocks with heavy retail trading but earn 6.6% among other stocks. Our results reconcile the literatures on the performance of retail investors, the predictive content of retail order imbalance, and attention-induced trading and returns. Smaller retail trades concentrate more on attention-grabbing stocks and perform worse.

Optimizing order policy and credit term for items with inventory-level-dependent demand under trade credit limit

Under the combined effects of inventory-level-dependent demand (ILDD) and trade credit, the retailer is able to order more quantities to stimulate market demand. However, from the supplier's perspective, two important issues are lacking sufficient attention. First, during the credit period, the retailer's higher order quantities imply increases in both the retailer's account payable and the supplier's opportunity cost of capital. Second, given the supplier's fixed production rate, the increased market demand may drive the capacity utilization to be variable. Thus, by formulating a supplier-dominated system, this paper incorporates trade credit limit (TCL) to address its effects on optimal policies vis-à-vis the item with ILDD. Specifically, three indicators can be proposed to reveal which type of financing policy the retailer should choose. Moreover, based on TCL, the supplier can effectively manage the retailer's order quantity and the corresponding account payable. Additionally, the retailer's maximum allowable order quantity is developed to ensure that the supplier can supply the retailer's order quantity on time. Furthermore, when the effects of ILDD become more significant, the manufacturer will reduce the maximum allowable order quantity to control the retailer's order incentive.

Investigation of the ordering behavior of a retailer in the revenue sharing and buyback contracts considering round number bias

This study investigates the ordering behavior of a boundedly rational retailer who purchases a product from a rational supplier under a revenue sharing or buyback contract. We develop a behavioral model considering random error, learning effect and anchoring bias using a logit choice framework which also includes a new behavioral paradigm known as round number bias. Experimental data and theoretical analysis are employed to demonstrate the differences between retailer order quantity and related optimal decisions using the suggested behavioral model. We found that round numbers are enticing anchors in retailer ordering decisions. The findings further reveal that the adaptive learning model better explains and predicts the retailer's ordering behavior than the time trend learning model. Moreover, the results indicate that the prediction accuracy of the model is higher for revenue sharing compared to buyback contracts.

Multi-store collaborative delivery optimization based on Top-K order-split.

Regarding the fulfillment optimization of online retail orders, many researchers focus more on warehouse optimization and distribution center optimization. However, under the background of new retailing, traditional retailers carry out online services, forming an order fulfillment model with physical stores as front warehouses. Studies that focus on physical stores and consider both order splitting and store delivery are rare, which cannot meet the order optimization needs of traditional retailers. To this end, this study proposes a new problem called the "Multi-Store Collaborative Delivery Optimization (MCDO)", in which not only make the order-split plans for stores but also design the order-delivery routes for them, such that the order fulfillment cost is minimized. To solve the problem, a Top-K breadth-first search and a local search are integrated to construct a hybrid heuristic algorithm, named "Top-K Recommendation & Improved Local Search (TKILS)". This study optimizes the search efficiency of the breadth-first search by controlling the number of sub-orders and improving the initial solution of the local search using a greedy cost function. Then achieve the joint optimization of order-split and order-delivery by improving the local optimization operators. Finally, extensive experiments on synthetic and real datasets validate the effectiveness and applicability of the algorithm this study proposed.

It Depends On When You Search

Existing studies have found that online search is a revealed measure for investor attention and a useful predictor of stock returns. We study the heterogeneity in retail investor attention by comparing search conducted on weekdays vs. weekends and investigate the price pressure channel and information processing channel for stock return predictability. According to the information processing channel, weekends afford retail investors more time for the intensive cognitive analysis necessary to make better predictions. Alternatively, weekend search might better capture the price pressure from retail investors’ trading activities. We provide empirical results that support the information processing channel. We first show that weekend search, rather than weekday search, predicts large-cap stock returns in both the cross-section and time series. Additionally, our findings on retail trading activity contradict the price pressure channel in that weekday search, rather than weekend search, leads to a subsequent retail order imbalance. Overall, our study contributes to the literature on the predictive power of online search on stock returns, which has mainly focused on the price pressure channel, which yields significant results for small-cap stocks only.

Omnichannel retail operations with ship-to-store and ship-from-store options under supply disruption

Omnichannel retailing strategies are widely used in practice and have been extensively studied in recent years, but few studies have explored omnichannel retailing operations in response to supply disruption in the post-pandemic era. To fill this gap, this study explores whether the adoption of omnichannel fulfillment options (i.e., ship-from-store and ship-to-store options) can mitigate the risk of supply disruption in a supply chain where a retailer orders products from a reliable supplier and a risky supplier, respectively. Under the omnichannel retailing strategy, the retailer’s order quantity from the risky supplier may increase or decrease while that from the reliable supplier may increase. Interestingly, it is possible to achieve a win–win–win outcome when the supply disruption risk is high and the market share of the channel offered by the risky supplier is low. Moreover, the entire supply chain benefits from the omnichannel retailing strategy even if it faces a high level of disruption risk.