Knowledge that Transforms

Problem definition: Last-mile delivery is one of the most challenging and costly facets of the supply chain. Using data from Rwandan public hospitals that transfuse blood, we examine whether and the extent to which the adoption of a technological innovation—delivery drones—for the last-mile delivery of blood products impacts the way hospitals manage the inventory of these products and the downstream health outcomes for patients. Methodology/results: To estimate causal effects, we exploit the staggered rollout of the drone delivery system and use a generalized difference-in-differences approach. We find that all hospitals, regardless of their distance to the drone port, substantially decrease their on-hand inventory of red blood cells, most notably for the most flexible product (i.e., type O blood). Hospitals also meaningfully reduce their wastage of red blood cells, especially for “in-between” products that are neither the most flexible nor the least flexible (i.e., type B blood). With respect to patient health outcomes, we document large reductions in in-hospital mortality for two health conditions where timely access to blood is a critical input for treatment: postpartum hemorrhage and trauma. Interestingly, this change is concentrated among hospitals closer to the drone port, and these are the same set of hospitals that increase the transfusion of blood products. This increase is especially pronounced among blood products that cannot be held in stock at the hospital, including platelets and fresh frozen plasma. Managerial implications: Our results highlight key considerations for decision makers allocating scarce resources to improve hospital operations and health outcomes. Given that health improvements are not realized by all hospitals, operational decisions—such as the number and location of drone ports—should ensure clinically meaningful delivery windows for time-sensitive medical care. History: This paper has been accepted in the Manufacturing & Service Operations Management Frontiers in Operations Initiative. Funding: This work was supported by the Wharton Dean’s Research Fund, Wharton Global Initiatives Research Program, Claude Marion Endowed Faculty Scholar Award, Mack Institute Research Fellowship, and Fishman-Davidson Center for Service and Operations Management. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2025.0055 .

Problem definition: We study a paradox in which decision-makers deviate downward from proposals of an automated store replenishment system after a stockout. We argue that censorship bias explains this curious ordering behavior and it has negative performance implications. We compare the effect of censorship bias to that of anchoring bias to understand the relative importance of censorship bias in retail practice. Understanding the impact of certain biases on decision-makers’ inventory replenishment decisions in the presence of an algorithm is imperative to maximize the benefit from decision-makers' discretionary power. Methodology/results: We analyze data on perishable products from an upmarket supermarket chain by employing exclusion restrictions in our recursive bivariate probit model to account for the endogeneity of deviations. We complement our analysis with endogenous switching regression and seemingly unrelated regression models to further test the robustness of our results. We find that after a stockout, decision-makers’ likelihood of deviating downward is higher, a behavior aligned with censorship bias. We find that anchoring bias is more powerful than censorship bias in predicting downward deviations. Regarding the performance implications, we show that censorship bias is more detrimental than anchoring bias in terms of increasing the likelihood of a new stockout. Therefore, we suggest that this insight into censorship bias can be used to distinguish uninformed downward deviations from the informed ones. Managerial implications: To suggest actionable policies, we collect more data to test the idea of blocking downward deviations when censorship bias is suspected. With this additional data analysis, we show that by blocking the downward deviations susceptible to censorship bias, retail managers can reduce self-inflicted stockouts with reasonable inventory cost implications.

Problem definition: A growing number of firms are encouraging consumers to participate in “bring-your-own-container” (BYOC) behavior in which consumers bring their own reusable packaging to purchase and consume products, thus reducing single-use packaging waste. In this paper, we study the environmental implications of a firm’s BYOC implementation when considering its disposable packaging choice and communication strategy. Methodology/results: We build a stylized model to study a firm’s joint decisions on BYOC, disposable packaging choice, and communication and their implications on the environment. Our main results follow. First, allowing BYOC reduces the firm’s incentive to make fraudulent green claims about its disposable product packaging; however, BYOC implementation may harm the overall environment while improving the firm’s profit, thereby creating a new form of greenwashing. Second, the adoption of third-party certification for green disposable packaging is an effective remedy to mitigate the negative environmental impact of BYOC. In addition, the environmental implications of adopting third-party certification (either voluntarily or because of government mandates) depend on the relationship between the environmental qualities of green disposable packaging and reusable packaging. Whereas it always benefits the environment when the firm’s green disposable packaging has better environmental performance, adopting certification may negatively impact the environment if consumers’ reusable packaging is greener. Furthermore, we find numerically that offering a price discount for BYOC may encourage the firm to adopt certification because of increased profitability, thereby leading to the aforementioned environmental implications. Managerial implications: We offer operational insights on how firms should make joint decisions on BYOC, disposable packaging choice, and communication. We also generate insights on how governments should regulate firms’ green claims when firms start to allow BYOC. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2021.0605 .

Problem definition: Online retailing platforms are increasingly relying on personalized recommender systems to help guide consumer choice. An important but understudied question in such settings is how many products to include in a recommendation set. In this work, we study how the number of recommended products influences consumers’ search and purchase behavior in an online personalized recommender system within a retargeting setting. Methodology/results: Via a field experiment involving 1.6 million consumers on an online retailing platform, we causally demonstrate that consumers’ likelihood of purchasing any product from the recommendation set first increases then decreases as the number of recommended products increases. Importantly, as much as 64% of the decrease in purchase probability (i.e., the choice overload effect) can be attributed to a decrease in consumers’ likelihood of starting a search (i.e., clicking on any recommended product). We discuss the possible behavioral mechanisms driving these results and analyze how these effects could be heterogeneous across different product categories, price ranges, and timing. Managerial implications: This work presents real-world experimental evidence for the choice overload effect in online retailing platforms, highlights the important role of consumer search behavior in driving this effect, and sheds light on when and how limiting the number of options in a recommender system may be beneficial to online retailers. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2022.0659 .

Problem definition: In this paper, we compare laissez-faire and mandatory authorization policy regimes for third-party remanufacturing. Under a laissez-faire policy, an independent remanufacturer (IR) chooses whether to get the original equipment manufacturer (OEM) authorization. Under a mandatory authorization policy, the IR is required to get OEM authorization and to pay the OEM a fee for every item remanufactured. Motivated by China’s regulatory journey that first mandated authorized remanufacturing and then moved to a laissez-faire policy, our goal is to understand which policy is better from the perspectives of different stakeholders. Methodology/results: We use a game-theoretic approach and consider a supply chain consisting of a supplier, an OEM, and an IR under the two policy regimes. Conventional wisdom suggests that the IR would be better off under the laissez-faire policy, but the OEM and the supplier would be better off under the mandatory authorization policy. However, we show that this conventional wisdom may not hold. For products with a low remanufacturing cost, all firms benefit from the mandatory authorization policy, whereas for products with a moderately high remanufacturing cost, all firms are better off under the laissez-faire policy. Further, mandatory authorization may outperform the laissez-faire policy in both economic and environmental dimensions. Managerial implications: Our findings reveal that seemingly advantageous policy regimes may backfire for firms. Therefore, before supporting such policies, the firms need to assess the strategic reactions of other firms and the potential impacts on their profits. Furthermore, a mandatory authorization policy can be beneficial in fostering the development of the remanufacturing sector for products with low remanufacturing costs. Nevertheless, it may also lead to an increase in the total environmental impact. Funding: The work of M. Jin was supported by the National Natural Science Foundation of China [Grants 72071020 and 72471038]. The work of Y. Zhou was supported by the National Natural Science Foundation of China [Grants 71971033 and 72371040] and the Fundamental Research Funds for the Central Universities [Grant 2024CDJSKPT14]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.0128 .

Problem definition: In response to the increased use of telehealth to replace traditional office visits with a physician, several U.S. states have recently adopted telehealth pay-parity policies. Such policies state that payers must reimburse healthcare providers for telehealth services at the same rate that would apply if those services had been provided in a traditional office setting. But health policy researchers have pointed out that telehealth may not be as effective as a traditional office visit for acute care. Specifically, telehealth is associated with increased probability of a subsequent office visit (a “duplicate visit”). We examine whether telehealth pay-parity policies are effective at improving access to acute care, and under what conditions. Methodology/results: We use a three-stage game-theoretic model to study the impact of telehealth pay parity. In the first stage, the payer sets a reimbursement policy for telehealth visits. In the second stage, a healthcare provider commits a portion of its capacity to telehealth, and in the third stage, patients arrive and choose between telehealth and office visits according to an equilibrium queueing network. We find structural results for the equilibria and characterize the equilibria in closed-form. When the chance of a duplicate visit is moderate (neither too high nor too low), pay parity leads providers to allocate too much capacity to telehealth, resulting in lower overall patient access than could be otherwise achieved. We characterize a reimbursement level that avoids this misalignment and maximizes patient access, which we show is less than parity. Managerial implications: The literature shows that patients receiving acute care via telehealth may be more likely to require a duplicate, in-person visit to resolve their health concern. In the fee-for-service environment that is common in the United States for acute care, duplicate visits resulting from telehealth lead to an incentive alignment problem because they generate extra work and provider revenue, without any corresponding increase in patient access. Legislating pay parity for telehealth can lead to providers committing more capacity to telehealth, which may not always be good. However, there is good news in that all parties (payers, providers, and patients) would be better off if duplicate visits could be decreased. Policy makers should understand these implications before enacting policies that affect reimbursements for telehealth. Funding: Support for this project was provided by a PSC-CUNY Award, jointly funded by The Professional Staff Congress and The City University of New York. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2022.0345 .

Problem definition: We study a feature-based pricing problem with demand censoring in an offline, data-driven setting. In this problem, a firm is endowed with a finite amount of inventory and faces a random demand that is dependent on the offered price and the features (from products, customers, or both). Any unsatisfied demand that exceeds the inventory level is lost and unobservable. The firm does not know the demand function but has access to an offline data set consisting of quadruplets of historical features, inventory, price, and potentially censored sales quantity. Our objective is to use the offline data set to find the optimal feature-based pricing rule so as to maximize the expected profit. Methodology/results: Through the lens of causal inference, we propose a novel data-driven algorithm that is motivated by survival analysis and doubly robust estimation. We derive a finite sample regret bound to justify the proposed offline learning algorithm and prove its robustness. Numerical experiments demonstrate the robust performance of our proposed algorithm in accurately estimating optimal prices on both training and testing data. Managerial implications: The work provides practitioners with an innovative modeling and algorithmic framework for the feature-based pricing problem with demand censoring through the lens of causal inference. Our numerical experiments underscore the value of considering demand censoring in the context of feature-based pricing. Funding: The research of E. Fang is partially supported by the National Science Foundation [Grants NSF DMS-2346292, NSF DMS-2434666] and the Whitehead Scholarship. The research of C. Shi is partially supported by the Amazon Research Award. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2024.1061 .

Problem definition: The meaning of life must surely be more about well-being than wealth, but what does that have to do with operations? Well-being encompasses a lot: Are we happy as individuals? Are groups treated fairly? Is society sustainable? Operations management has many impacts on well-being at each of these levels, some more obvious than others. Methodology/results: This MSOM Fellow forum article offers a wide-ranging exploration of linkages between operations and well-being. It organizes “operations” into five broad areas: pace and productivity, predictability and probability, process and prevention, performance and payment, and pollution and protection. For each of those, it explores what makes individuals (un)happy, what is fair, and what is sustainable. Managerial implications: It concludes with 7 recurrent themes for research in Operations Management.

Problem definition: The loyalty penalty refers to a pricing strategy where companies charge higher prices to loyal customers for exploitation while offering lower prices to nonloyal customers for attraction. To address this unfair business practice, various regulatory agencies, such as the Competition and Markets Authority and the Financial Conduct Authority in the United Kingdom, have proposed or implemented regulations aimed at promoting fairness in pricing. In this study, we analyze the impact of such regulations on both firms and consumers. Methodology/results: We develop a stylized model to investigate duopoly competition in two symmetric markets, where consumers exhibit loyalty to different firms in each market. The regulatory intervention mandates that the price difference between the two markets, set by each firm, must not exceed a certain threshold. Our analysis reveals an intriguing interaction between market competition and price fairness regulation. When competition is intense, fairness regulation can alleviate cutthroat competition between firms, resulting in Pareto improvements compared with a scenario with no regulation. On the other hand, when competition is weak, fairness regulation can further enhance firms’ existing monopoly power, potentially leading to collusive high prices that are detrimental to consumers and society. We also consider several extensions to enrich our findings, including fairness regulation on relative price discounts, asymmetric markets, and a two-pronged policy regulating the price gap and price cap. Managerial implications: Our study reveals the economic consequences of price fairness regulations; firms can benefit by pricing fairly, but well-intended fairness requirements may have adverse effects on consumers. Funding: P. Gao is supported by the National Natural Science Foundation of China [Grants 72201234 and 72192805], the Hong Kong Research Grants Council (the Collaborative Research Fund) [Grant C6032-21G], and the Guangdong Provincial Key Laboratory of Mathematical Foundations for Artificial Intelligence [Grant 2023B1212010001]. Y.-J. Chen acknowledges financial support from the Hong Kong Research Grants Council [Grants CRF HKUST C6020-21GF, GRF 16500821, and GRF 16501722]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2022.0552 .

Problem definition: Contrary to traditional queueing theory, recent field studies in B2C services indicate that pooled queues may be less efficient than dedicated queues. Methodology/results: We use two online experiments in the healthcare delivery context to replicate this finding and assess the interplay of servers’ customer ownership and queue length awareness as potential underlying mechanisms. We operationalize customer ownership as the extent to which servers feel ownership toward their customers and queue length awareness as the extent to which servers are able to accurately quantify their number of customers. We find that, following a change in queue configuration, dedicated queues outperform pooled queues with respect to processing speed without sacrificing quality. The reduction in speed is partially mediated by the servers’ queue length awareness and partially suppressed by their ownership of customers in queue. The former is because servers turn out to be less likely to underestimate their load, which makes them work faster. The latter is because ownership of customers in queue may distract servers from the customer in service. When the queue configuration changes from a dedicated to a pooled one, the shorter processing times and higher levels of queue length awareness persist across the change, unlike the higher ownership of customers in the queue. Managerial implications: In discretionary service settings, switching to a dedicated queue is often beneficial in terms of operational performance, partly because the increased queue length awareness motivates servers to work faster; however, the increased degree of customer ownership of those in queue may distract them and result in a slowdown. Funding: This work was supported by the Wharton Behavioral Lab, the Claude Marion Endowed Faculty Scholar Award, the Wharton-INSEAD Alliance, and the Wharton Dean’s Research Fund. Supplemental Material: The online appendices are available at https://doi.org/10.1287/msom.2023.0202 .