This paper examines the role of information sharing in supply chain management, facilitated by recent technological advancements like cloud-based systems and smart contracts. We analyze the optimal information disclosure policy in a two-tier supply chain with a retailer and a supplier, highlighting how the alignment of incentives affects the level of disclosure. In contrast to the vast literature on information sharing in supply chain management, our study is the first one focusing on settings where a retailer can commit to her information disclosure policy. Thus, our findings show that a well-designed information sharing policy can significantly benefit the retailer, offering new perspectives on supply chain information management.

We explore the challenges decision makers face when assessing the quality of machine learning recommendations compared to their judgment. Our analysis reveals that decision makers may remain uncertain about the superiority of algorithmic recommendations, leading to frequent overruling or may incorrectly believe the machine is better. We identify conditions under which these learning failures occur, contributing to the literature on human-machine interaction and decision-making. 

This study focuses on how major tech companies manage information disclosure when launching new products. We use a game-theoretic model of Bayesian persuasion to show how firms tailor their strategies based on market dynamics such as polarized customer beliefs and information acquisition capabilities. We characterize conditions under which companies may fully disclose, exaggerate, downplay, or withhold information. Our work provides a framework for understanding strategic information disclosure depending on market heterogeneity and customers’ access to external information.

Our research examines how governments can use strategic communication to influence public compliance with lockdowns and confinement measures during a pandemic. By applying an information design framework, we reveal that governments might adjust the disclosure of an epidemic’s severity based on economic and health priorities. Notably, we find that economic inequality affects how informative these communication strategies are. This study underscores the complexity of public health communication and its implications for policy effectiveness.

In this paper, we investigate the challenges firms face in implementing the AI Flywheel effect, where machine learning algorithms improve with more data. We explore the interactions between data acquisition, pricing, and contracting, highlighting how these factors can create decision distortions. Our results show that firms could enhance profits by carefully managing data collection, but excessive data acquisition might reduce profitability. This research contributes to the understanding of data-driven business strategies such as pricing of AI products and contracting for AI services.

This study explores the complex ecosystem of display advertising, where intermediaries acquire ad impressions for advertisers. We characterize bidding strategies and selling mechanisms in the intermediary network. Our findings reveal that network structure significantly influences intermediary profits and economic alignment. Additionally, we examine the impact of tree structures on seller revenues and potential incentives for intermediaries to merge. This research provides valuable insights into the strategic dynamics of online advertising networks.

In this paper, we address the challenge of efficiently allocating scarce resources in settings where monetary transactions are impractical, such as antipoverty programs, healthcare and cloud computing. We design a dynamic mechanism (a long-term allocation strategy without monetary transfers) that promotes truthful reporting from requesters by adjusting future resource availability based on current disclosures, and also approaches the most efficient allocation as requesters become more patient. This work contributes to the understanding of non-monetary resource allocation systems. Furthermore, in the case of two requesters no other mechanism can converge faster to first best.

To place an emphasis on profound relations among airline schedule planning problems and to mitigate the effect of unexpected delays, we integrate schedule design, fleet assignment and aircraft routing problems within a daily planning horizon while passengers’ connection service levels are ensured via chance constraints. We propose a nonlinear mixed integer programming model due to the nonlinear fuel consumption and CO2 emission cost terms in the objective function, which is handled by second order conic reformulation. The key contribution of this study is to take into account the cruise time control for the first time in an integrated model of these three stages of airline operations. 

Deep tech innovation involves significant advancements in both scientific and business domains, often necessitating joint ventures between co-founders with complementary expertise. This study examines the innovation search process, focusing on how the maturity of initial ideas, search costs, and innovation uncertainty affect decision-making. Our findings reveal that co-founders’ willingness to pursue innovation is non-linear concerning the maturity of their ideas. Interestingly, a more mature idea may re-motivate involvement even after abandoning less developed concepts. This work highlights the critical role of balancing scientific and business dimensions in successful deep-tech ventures and suggests that replacing a co-founder with a more mature idea may sometimes backfire rather than help innovation.

Startup firms often lack separate marketing and operations departments, requiring them to make integrated decisions across these areas without complete data. This paper explores a scenario where a firm must simultaneously learn about customer preferences and determine pricing, advertising, and inventory strategies over multiple periods. Given the computational challenges of finding the optimal policy, we identify a family of policies that enable rapid learning and propose an easy-to-implement, asymptotically optimal solution. Our results indicate that understanding advertising responses is crucial for effectively determining pricing strategies, ensuring minimal profit loss compared to an ideal scenario with perfect information.