Knowledge that Transforms

Near‐constant Internet access through desktop or mobile devices has turned self‐service support forums into the first port of call for users seeking to troubleshoot product or service issues. The firms providing these products and services also benefit from this trend since it reduces user support costs by diverting service requests away from costlier support channels, such as help desks. For the continued success of such a forum, however, the managing entity must ensure that users receive timely solutions to their inquiries quickly and regularly. We develop a mathematical model of a user forum's operations to obtain a “white box” view of a user forum and reveal the support system's dynamics. Then, using a large and comprehensive dataset of questions and answers from Apple's iPhone user forum, we empirically estimate the forum's performance to validate the predictions of the mathematical model. Our results demonstrate that the predictions closely match the forum's actual performance, with an error of less than 10%. We then propose and analyze an optimal threshold policy that boosts a thread to rekindle user interest and demonstrate the benefit of our intervention policy in managing the iPhone forum.

We consider a periodic review single‐item inventory model under stochastic demand. Every m periods, in the regular order period, fixed order costs are K . In the periods in‐between, the intraperiods, higher fixed order costs of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:semantics definitionURL="" encoding=""> <mml:mrow> <mml:mi>L</mml:mi> <mml:mo>></mml:mo> <mml:mi>K</mml:mi> </mml:mrow> <mml:annotation encoding="">L>K</mml:annotation> </mml:semantics> </mml:math> apply. The literature on optimal inventory policies under fixed order costs does not account for these time‐dependent fixed order costs. By generalizing existing proofs for optimal inventory policies, we close this gap in inventory theory. The optimal inventory policy is complex in the regular order period and in the intraperiods, a period‐dependent <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:semantics definitionURL="" encoding=""> <mml:mrow> <mml:mo stretchy="false">(</mml:mo> <mml:mi>s</mml:mi> <mml:mo>,</mml:mo> <mml:mi>S</mml:mi> <mml:mo stretchy="false">)</mml:mo> </mml:mrow> <mml:annotation encoding="">(s,S)</mml:annotation> </mml:semantics> </mml:math> policy is optimal. We describe and prove this optimal policy based on the notion of K ‐convexity and the optimal ordering behavior in the presence of non– K ‐convex cost functions. In a numerical study, we find that a major driver of the optimal policy is a forward‐buying effect that shifts the probability of ordering from the intraperiods to the regular order period. The cost differences between the optimal and a pure period‐dependent <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:semantics definitionURL="" encoding=""> <mml:mrow> <mml:mo stretchy="false">(</mml:mo> <mml:mi>s</mml:mi> <mml:mo>,</mml:mo> <mml:mi>S</mml:mi> <mml:mo stretchy="false">)</mml:mo> </mml:mrow> <mml:annotation encoding="">(s,S)</mml:annotation> </mml:semantics> </mml:math> policy are, however, small.

We develop a general framework for selecting a small pool of candidate solutions to maximize the chances that one will be optimal for a combinatorial optimization problem, under a linear and additive random payoff function. We formulate this problem using a two‐stage distributionally robust model, with a mixed 0–1 semidefinite program. This approach allows us to exploit the “diversification” effect inherent in the problem to address how different candidate solutions can be selected to improve the chances that one will attain a high ex post payoff. More interestingly, using this distributionally robust optimization approach, our model recovers the “evil twin” strategy, well known in the field of football pool betting, under appropriate settings. We also address the computational challenges of scaling up our approach to construct a moderate number of candidate solutions to increase the chances of finding one that performs well. To this end, we develop a sequential optimization approach based on a compact semidefinite programming reformulation of the problem. Extensive numerical results show the superiority of our approach over existing methods.

We study capacity sizing of park‐and‐ride lots that offer services to commuters sensitive to congestion and parking availability information. The goal is to determine parking lot capacities that maximize the total social welfare for commuters whose parking lot choices are predicted using the multinomial logit model. We formulate the problem as a nonconvex nonlinear program that involves a lower and an upper bound on each lot's capacity, and a fixed‐point constraint reflecting the effects of parking information and congestion on commuters' lot choices. We show that except for at most one lot, the optimal capacity of each lot takes one of three possible values. Based on analytical results, we develop a one‐variable search algorithm to solve the model. We learn from numerical results that the optimal capacity of a lot with a high intrinsic utility tends to be equal to the upper bound. By contrast, a lot with a low or moderate‐sized intrinsic utility tends to attain an optimal capacity on its effective lower bound. We evaluate the performance of the optimal solution under different choice scenarios of commuters who are shared with real‐time parking information. We learn that commuters are better off in an average choice scenario when both the effects of parking information and congestion are considered in the model than when either effect is ignored from the model.

This paper proposes price cannibalization as a growth strategy despite prior findings that suggests avoiding it. We focus on a multiclass, capacity‐constrained pricing problem in which each of the product classes has a price range. Specifically, we examine the effects of price range overlaps and introduce it as a revenue‐maximizing pricing strategy. Price cannibalization happens when sales in some product classes decrease due to the existence of overlaps between the price ranges. We employ a multimethod approach. First, we define a Markov decision problem to obtain the revenue‐maximizing strategy in a two‐class sales scenario. We show that price range overlaps are part of the optimal strategy. Second, we collect multichannel data from a European storage company to examine how price range overlaps impact a customer's purchase decisions. The results show that the existence of price range overlaps leads to cannibalization, but increases spending and improves conversion. Finally, we use simulations to compare several pricing strategies and demonstrate the long‐term effects of using price range overlaps in pricing algorithms in complex situations. Our findings suggest that using price range overlaps, though leads to cannibalization, actually helps companies avoid spoilage and early sellouts, leading to better capacity utilization and higher revenue.

We study inventory control with volume flexibility: A firm can replenish using period‐dependent base capacity at regular sourcing costs and access additional supply at a premium. The optimal replenishment policy is characterized by two period‐dependent base‐stock levels but determining their values is not trivial, especially for nonstationary and correlated demand. We propose the Lookahead Peak‐Shaving policy that anticipates and peak shaves orders from future peak‐demand periods to the current period, thereby matching capacity and demand. Peak shaving anticipates future order peaks and partially shifts them forward. This contrasts with conventional smoothing, which recovers the inventory deficit resulting from demand peaks by increasing later orders. Our contribution is threefold. First, we use a novel iterative approach to prove the robust optimality of the Lookahead Peak‐Shaving policy. Second, we provide explicit expressions of the period‐dependent base‐stock levels and analyze the amount of peak shaving. Finally, we demonstrate how our policy outperforms other heuristics in stochastic systems. Most cost savings occur when demand is nonstationary and negatively correlated, and base capacities fluctuate around the mean demand. Our insights apply to several practical settings, including production systems with overtime, sourcing from multiple capacitated suppliers, or transportation planning with a spot market. Applying our model to data from a manufacturer reduces inventory and sourcing costs by 6.7%, compared to the manufacturer's policy without peak shaving.

Millions of workers in India who migrated to cities for employment have transited back to seek refuge in their home villages, causing disruptions in both cities and villages. This type of mass flow of migrants in transit represents a humanitarian crisis. Understanding migrant flow patterns and ways to ameliorate the conditions for migrants in transit is critical to managing the humanitarian crisis. In this study, we develop a model that examines the influence of migrant networks, inter‐organizational collaboration, and environmental uncertainty on locational advantage, which, in turn, predicts migrant flow patterns. This study contributes to the humanitarian operations management and migration literatures by uncovering how migrant networks and inter‐organizational collaboration help provide access to humanitarian resources. Additional new findings of this study include uncovering different classes of migrants with their respective flow patterns and the role of collaboration along migration paths. The study also uncovers how travel constraints increase the duration of transit and the importance of point‐to‐point transfers to avoid congregation at transit hubs. Furthermore, findings from this research provide insights on how long‐term humanitarian support to migrants through inter‐organizational collaboration morphs to short‐term aid in the event of a crisis.

Patient‐reported outcomes (PROs) play an increasingly important role in medical decision making. Yet, patients whose objectives differ from their physician's may strategically report symptoms to alter treatment decisions. For example, athletes may underreport symptoms to expedite return‐to‐play (RTP) from sports‐related concussion (SRC). Thus, clinicians must implement treatment policies that mitigate the Price of Naïveté , that is, the reduction in health outcomes due to naïvely believing strategically reported symptoms. In this study, we analyze dynamic treatment cessation decisions with strategic patients. Specifically, we formulate the Behavior‐Aware Partially Observable Markov Decision Process (BA‐POMDP), which optimizes the timing of treatment cessation decisions while accounting for known symptom‐reporting behaviors. We then analytically characterize the BA‐POMDP's optimal policy, leading to several practical insights. Next, we formulate the Behavior‐Learning Partially Observable Markov Decision Process (BL‐POMDP), which extends the BA‐POMDP by learning a patient's symptom‐reporting behavior over time. We show that the BL‐POMDP is decomposable into several BA‐POMDPs, allowing us to leverage the BA‐POMDP's structural properties for solving the BL‐POMDP. Then, we apply the BL‐POMDP to RTP from SRC using data from 29 institutions across the United States. We estimate the Price of Naïveté by comparing the BL‐POMDP to naïve benchmark policies. Accordingly, the BL‐POMDP reduces premature RTP by over 44% and provides up to 3.63 additional health‐adjusted athletic exposures per athlete compared to current practice. Overall, changing the interpretation of reported symptoms can better reduce the Price of Naïveté over adjusting treatment cessation thresholds. Therefore, to improve patients' health outcomes, clinicians must understand how strategic behavior manifests in PROs.

We study the optimal design of a professional service in a mixed market of customers with heterogeneous skills and capabilities of using such service. Expert customers can avail of the service on their own, whereas amateur customers find it challenging to deploy the service and can only procure the service through an intermediary who resolves the technical issues. We develop a model that captures the essential trade‐offs in such settings: heterogeneity in customer expertise, decentralization between a service provider and intermediary, and congestion due to limited capacity. We analyze how customer expertise differences drive the equilibrium outcomes under various pricing and priority schemes. We find that a sufficient base of amateur customers allows expert customers to “free‐ride” under single pricing. Price discrimination can fully allay such free‐riding, but it may drive prices downward. Price discrimination also favors expert customers under the First‐Come‐First‐Served (FCFS) policy, but such preference is generally reversed under prioritization. Specifically, prioritizing amateur customers can bring revenue and welfare gains relative to the FCFS policy and a policy that prioritizes expert customers. Our results offer normative guidelines for managing professional services, clarifying regimes for price and priority discrimination, along with revenue and welfare implications.

We propose and analyze a novel framework to understand the role of noninstrumental information sharing in service operations management, that is, information shared by the firm not to affect consumers' actions, but to better manage their experience in the firm's process. The operations of the firm are organized as a process , consisting of a sequence of tasks, each of random duration. The firm shares real‐time information with the consumer about the progress of their flow unit in the firm's process via a process tracker . The consumer is delay‐sensitive and experiences gain–loss utility (loss aversion and diminishing sensitivity) over time due to changes in beliefs about anticipated delay, as he awaits completion of the process. We analyze when providing such real‐time progress information via process trackers help, or can possibly hurt a consumer. Our work draws upon the recent literature on belief‐based/news utility in Economics. We find that in the presence of loss aversion alone, not sharing progress information is beneficial. In the presence of loss aversion and diminishing sensitivity, if low delays are likely, then sharing information is beneficial; otherwise, not sharing information is preferred. Our findings inform a service firm's post‐sales transparency strategy.