Papers

• Information Design of a Delegated Search, with Yangge Xiao and Zhenyu Hu. (Under revision at Management Science)

  • We formulate the dynamic information design of a delegated search and fully characterize its optimal information policy, which is completely prescribed by a sequence of deterministic acceptance standards. The agent is recommended to continue the search if the current termination payoff fails to meet that period’s standard. For searches like talent recruitment, where outcomes are not recallable, acceptance standards should be gradually loosened as the search progresses so as to back-load the search incentives. On the other hand, for innovation-driven searches, like R&D or academic research, where search outcomes accumulate, search incentives should be front-loaded by adopting a hands-off approach early in the search, and only nudging the search later on.

  • Runner-up, 2023 Decision Analysis Society (DAS) Student Paper Award

• The Impact of Climate Change: An Empirical Analysis of Smart Thermostat Data, with Michael Blair and Saed Alizamir. (Under revision at Management Science)

  • Leveraging a high-frequency smart thermostat data set, we empirically examine how households' energy consumption decisions respond to exogenous weather in the long and short run. As a policy insight, energy conservation should dissuade households from relying on long-term automation, while incentivizing short-term adherence to pre-programmed decisions.

• Recommender Systems under Privacy Protection, with Can Kücükgül and Özalp Özer. (Under revision at Management Science)

  • Personal preferences over different product offerings are a basic constituent of consumer privacy. We study how a profit-driven online platform designs its recommender system in response to different regulatory levels of privacy protection. We find that the opt-out privacy protection does not hinder the platform’s profitability nor the user’s surplus when compared to unprotected privacy. However, the self-disclosure option, which offers users the highest autonomy over their privacy, may lead to algorithmic discrimination, whereby the disadvantaged minority in the society are restricted or deprived of access to potential valuable opportunities.

• Audit and Remediation Strategies in the Presence of Evasion Capabilities, with Francis de Véricourt and Peng Sun. (Minor revision at Operations Research)

  • In a continuous-time dynamic mechanism design framework, we formulate a principal’s problem of seeking to uncover and remedy an issue that occurs to an agent at a random point in time. Only the agent observes the issue’s occurrence and can evade the principal’s audits. We find that the principle should implement a cyclic auditing and remedial cost-sharing mechanism. Importantly, each auditing cycle features a deterministic no-audit period followed by a random exponential audit.

• Avoiding Fields on Fire: Information Dissemination Policies for Environmentally Safe Crop-Residue Management, with Mehdi H. Farahani, Milind Dawande, and Ganesh Janakiraman, Management Science, forthcoming.

  • Burning crop residue in harvested fields to prepare land for sowing a new crop is well-recognized as a significant contributor to CO2 and black-carbon emissions. A specific agricultural machine called Happy Seeder has emerged as the most effective and profitable alternative to open burning.We study how the government can use effective information-disclosure policies in the operation of Happy Seeders to minimize agricultural open burning.

  • Honorable Mention, 2022 Best Paper Award, Humanitarian Operations & Crisis Management Track of POM Society

• A Stationary Infinite-Horizon Supply Contract under Asymmetric Inventory Information, with Alain Bensoussan and Suresh Sethi, Operations Research, forthcoming.

  • We consider a decentralized supply chain where a supplier sells goods to a retailer facing a general random demand over an infinite horizon. The retailer satisfies the demand to the extent of the inventory on hand. The retailer has private information about his inventory in each period, and the supplier offers him a dynamic supply contract menu to account for such information asymmetry. The supplier’s contract design problem can be formulated as a functional optimization with the retailer’s incentive compatibility constraint expressed as a functional Bellman equation. Mathematically, the problem belongs to calculus of variations, and requires us to develop powerful new methodologies by replacing the usual concept of gradient with that of Gateaux derivative.

• Engineering Social Learning: Information Design of Time-Locked Sales Campaigns for Online Platforms, with Can Kücükgül and Özalp Özer, Management Science, 68(7): 4899-4918, 2022.

  • On an Open Question in Kücükgül, Özer, and Wang (2022), with Yang Bo, technical note.

  • Second Prize, 2020 Jeff McGill Student Paper Award by INFOMRS Revenue Management & Pricing Section

  • Video presentation by Can Kücükgül

  • Using a dynamic Bayesian persuasion framework, we formulate and study how a platform should optimally design its dynamic information provision strategy to maximize its expected revenue for a time-locked sales campaign. We make three contributions. First, we establish an equivalent reformulation of the platform’s information design problem as an LP by significantly reducing the dimensionality of the platform’s message space and proprietary history. Second, we develop an easy-to-implement, simple-to-prescribe, and near-optimal policy, which we characterize analytically. Finally, we demonstrate the value of information design against some benchmark policies commonly used in practice and offer prescriptive managerial insights.

• Informing the Public about a Pandemic, with Huseyin Gurkan and Francis de Véricourt, Management Science, 67(10): 6358-6377, 2021.

  • Press Coverage by United Nations, ESMT, EurekAlert, World News Monitor, UTD News Center

  • The COVID-19 pandemic has highlighted the importance of properly informing the population about the epidemic when implementing governmental confinement measures. Indeed, the efficacy of these restrictions relies heavily on public compliance. Our work sheds light on how governments should inform the population to induce such social distancing behavior. Using an information design framework, our model explicitly account for the health-economy tradeoffs, the strategic externalities of individual compliance behaviors, and the economic inequality of the population.

• The Informational Role of Buyback Contracts, with Haresh Gurnani and Upender Subramanian, Management Science, 67(1): 279-296, 2021.

  • We examine the informational role of buyback contracts in credibly signaling either a manufacturer’s reliability of honoring the buyback commitment, or her product’s market potential. We find that these two situations form contrasting buyback designs: a manufacturer must distort the wholesale and returns prices downward to signal higher reliability and upward to signal higher market potential. The returns price emerges as a more efficient signaling instrument and reverses the direction of wholesale price distortion from what is necessary if the wholesale price alone is used to distort the returns cost.

• Procurement with Cost and Non-Cost Attributes: Cost-Sharing Mechanisms, with Shivam Gupta, Milind Dawande, and Ganesh Janakiraman, Operations Research, 69(5): 1349-1367, 2021.

  • Selected to present at M&SOM 2018 Conference – SIG Meeting on Supply Chain Management

  • We study an easy-to-implement class of cost-sharing mechanisms as the solution to a prevalent procurement problem, whereby the buyer faces two-dimensional private information: cost and non-cost attributes. We show how the cost-sharing fraction strikes the fundamental tradeoff between allocative inefficiency versus information rent. We characterize the optimal cost-sharing fraction in closed form and offer prescriptive guidelines on the choice of this fraction purely based on the second-moment information of the buyer’s belief distribution. We demonstrate, both theoretically and numerically, that the best cost-sharing mechanism is near-optimal.

• Warning Against Recurring Risks: An Information Design Approach, with Saed Alizamir and Francis de Véricourt, Management Science, 66(10): 4612-4629, 2020. (Former title: Design of Public Warning Systems)

  • Winner of 2018 Best Paper Award, Humanitarian Operations & Crisis Management Track of POM Society

  • Press Coverage by Yale Insight and Forbes

  • We study a novel information design problem addressing the issue of sending warning messages against recurring risks, a challenge particularly prevalent in global health and disaster management arena. Our framework accounts for dynamic learning effects and which we solve in closed-form. We uncover nuanced effects that induce a warning sender to strategically exaggerate or downplay a risk when alerting an uninformed stakeholder against an upcoming threat.

• Designing Sustainable Products under Co-Production Technology, with Yen-Ting Lin and Haoying Sun, M&SOM, 22(6): 1181-1198, 2020.

  • Motivated by emerging business practice, such as Taylor Guitars, our study informs firms when and how to leverage a co-product’s environmental value through product positioning decisions (i.e., quality and price). We find that, for intermediate material cost and lower green consumer’s environmental valuation, the firm should position the co-product without extracting its environmental value from green consumers. Otherwise, the firm should position the co-product to extract all consumer surplus by charging an additional price premium. In the latter case, the firm may strategically abandon some traditional consumers, who are, in fact, willing to purchase the more expensive traditional product, and leave their demand unfulfilled. Our research also yields important environmental and policy implications regarding conservation of scarce resources.

• Signaling Product Quality through a Trial Period, with Gülru Özkan-Seely, Operations Research, 66(2): 301-312, 2018.

  • Press Coverage by Newswise.com

  • A product trial period offers a learning channel, through which consumers can privately update their heterogeneous assessment of the product. This enables the trial length as a novel signal of the seller’s unobservable product quality in addition to the conventional price signal. This paper formalizes this idea in a continuous-time Bayesian learning model.

  • Note: Entrant’s Product Quality Signaling through a Trial Period in a Competitive Market

  • This note provides a self-contained analysis of an entrant-incumbent competition as an extension of the monopolistic trial period signaling model examined by the paper above.

• Resource Allocation under Demand Uncertainty and Private Information, with Alexandre Belloni and Giuseppe Lopomo, Management Science, 63(12): 4219-4235, 2017.

  • Our findings call attention to the strategic benefits that may arise in the presence of multilateral private information from consolidating the bargaining power to the sector that is directly exposed to demand uncertainty. A novel type of distortion, which we term as “overshooting,” emerges. These results resonate with the significantly different performances of the US agricultural produce and flu vaccine markets.

• Inducing Environmental Disclosures: A Dynamic Mechanism Design Approach, with Peng Sun and Francis de Véricourt, Operations Research, 64(2):371-389, 2016.

  • Winner of 2018 INFORMS ENRE Best Publication Award for Environment and Sustainability

  • Selected to present at 2nd Annual Workshop on Sustainable Supply Chain Management

  • Selected to present at M&SOM 2015 Conference – SIG Meeting on Sustainable Operations

  • Selected to present at 7th Annual ARCS Research Conference

  • Press Coverage by Inside JSOM and Duke Fuqua School of Business

  • Using the framework of dynamic mechanism design with state verification in continuous time, this paper studies how to design an efficient voluntary disclosure program (e.g., US EPA’s Audit Policy) that combines the two regulatory instruments: inspection and fiscal reward. We obtain a complete analytical characterization of the optimal regulation policy.

• Network Routing in Insurgency: An Adversarial Risk Analysis Framework, with David Banks, Naval Research Logistics, 58(6): 595-607, 2011.

• Adversarial Risk Analysis: Borel Games, with David Banks and Francesca Petralia, Applied Stochastic Models in Business and Industry, 27: 72-86, 2011.

  • The above two articles develop the adversarial risk analysis (ARA) framework and apply it to the classical Borel games and national defense setting.

• Multi-dimensional Mechanism Design: Finite Dimensional Approximations and Efficient Computation, with Alexandre Belloni and Giuseppe Lopomo, Operations Research, 58(4):1079-1089, 2010.

  • Funded by NSF

  • This paper considers how to allocate a single object of different quality dimensions to multiple buyers who have privately known valuation for each quality dimension. We find an efficient algorithm to solve the problem numerically and identify “exclusive buyer mechanism” as a near-optimal implementation of the optimal mechanism.

• Decentralized Resource Allocation to Control an Epidemic: A Game Theoretic Approach, with Francis de Véricourt and Peng Sun, Mathematical Biosciences, 222(1):1-12, 2009.

  • This paper studies the decentralized decision of allocating medical resources within and across borders by two self-interested countries that face the threat of an epidemic disease outbreak.

Technical Reports

• Implementation of Reduced Form Auction for Asymmetric Bidders with Continuous Types, 2008.

• A Simple Application of Seminorm Representation in Dynamic Programming, 2008.

Works in Progress

• Strategic Role of Service Quality Monitoring, with Aleda Roth and Qiong Chen.