Preprints/Working Papers
1. Potty Parity: Process Flexibility via Unisex Restroom
Setareh Farajollahzadeh, Ming Hu
Status: Under Second Round of Revision at Management Science
Keywords: Queueing
Awards: Finalist, CORS Best Student Paper, Applied Probability and Queueing 2023
Abstract: We address unequal restroom access for women and LGBTQ+ individuals, known as the ``potty parity" problem. We propose a utility model where users consider gender identity, wait time, and safety concerns when choosing restrooms. We evaluate different layouts' efficiency by total utilities (totalitarian principle) and assess their fairness using minimum utility gain (Rawlsian fairness) and the gap between maximum and minimum gains (distributive fairness). While it may initially seem intuitive to assume that converting all restrooms to unisex facilities would be efficient due to the pooling of servers and increased flexibility and fairness due to all users standing in the same line, our findings demonstrate that this design can be neither efficient nor fair. In contrast, we show that converting some of the men's restrooms to unisex facilities can enhance both efficiency and fairness of access. This highlights that a moderate level of flexibility can surpass a fully flexible system. Moreover, conventional wisdom suggests that removing a unit of restroom from the men's room would negatively impact users from the men's side. However, our analysis reveals a counterintuitive result that such a change can lead to a Pareto improvement, benefiting all users involved. We also analytically explore additional benefits of unisex restrooms under different user behaviors and situations and present numerical results to support our findings.
2. Sharing Newsboys
Setareh Farajollahzadeh, Ming Hu
Status: Major Revision at Operations Research
Keywords: Network games
Awards: Selected for presentation at MSOM SIG 2023
Abstract: We consider a network of socially connected newsvendors facing random demand for a product who need to commit to a stocking level before demand realizes. A newsvendor can share her ex post excess stock to fulfill the unsatisfied demand of a connected newsvendor. The amount of shared supply that a newsvendor anticipates receiving from her network is affected by two factors: sharing magnitude and tie strength. Sharing magnitude (resp., tie strength) measures the portion of excess stock that a newsvendor will share (resp., the likelihood that a newsvendor will share her excess supply) with a neighboring newsvendor. We adopt a Bayesian game framework with incomplete information about the network structure, where a newsvendor has private information about the number of connections she has (as her type) but does not know her neighbors' types, which she believes are consistent with a network's known degree distribution. First, we demonstrate that with more sharing activity (i.e., greater sharing magnitude or stronger social ties) within a fixed network, all newsvendors decrease their stocking levels regardless of their types, which implies that the total consumption level drops. Second, we show that when tied with the number of connections a newsvendor has, the sharing magnitude has a first-order effect on the mean of the shared supply, while the social tie has a second-order effect on the variability of the shared supply. As the degree distribution of the network increases in the sense of usual stochastic dominance, we show that the two factors may have opposite effects on the equilibrium stocking levels. The effect of sharing magnitude is to increase the equilibrium stocking levels. But the effect of tie strength is such that for a high-fractile product, the population's expected consumption level increases, while it is the other way around for a low-fractile product. Lastly, we extend the supply-sharing base model to complete network information under specific networks and to demand sharing, where unsatisfied demand at one newsvendor can be referred to a neighbor in her network.
3. Learning Customer Preferences from Bundle Sales Data
Ningyuan Chen, Setareh Farajollahzadeh, Guan Wang
Status: Submitted
Keywords: EM algorithm, bundle, demand estimation
Awards: Winner of CORS Best Student Paper, Open Category 2023
Abstract: This paper studies estimation problem of customer preferences from bundle sales data. Product bundling is a common selling mechanism used in retails. To set profitable bundle selection and prices, the seller needs to learn the distribution of consumers' valuations for individual products from the transaction data. When customers purchase bundles or multiple products, classical methods such as discrete choice models cannot be used to estimate customers' valuations. In this paper, we propose an approach to learn the distribution of consumers' valuations toward the products using bundle sales data. Our approach is to define a utility model for customer choices and estimate the parameters of a valuation distribution that maximizes the likelihood of observing the transaction data. Our approach reduces this problem to an estimation problem where the samples are censored by polyhedral regions on the valuation space of customers. Using the EM algorithm and Monte Carlo simulation, our approach can recover the distribution of consumers' valuations. We extend the framework to allow for unobserved no-purchases and clustered market segments. In addition, we provide theoretical results on the identifiability of the probability model and the convergence of the EM algorithm. Moreover, the performance of the approach is also demonstrated numerically with synthetic and real datasets. This study demonstrates the need and challenge for retailers to leverage the transaction data of bundle sales to learn customers' preferences. The proposed algorithm can be used efficiently in practice to achieve the goal.
4. Simultaneous vs. Sequential Product Release
Hojat Abdollahnejad, Ningyuan Chen, Setareh Farajollahzadeh, Ming Hu
Status: Major Revision at Manufacturing & Service Operations Management
Keywords: Learning, Reviews, Buzz economy
Abstract: We investigate the strategic release choices of a serial content creator with an unknown attraction level to customers before its debut. Employing an analytical framework, we examine scenarios in which the content creator must choose a release strategy, factoring in price flexibility and customer learning structures. The creator decides between releasing content simultaneously with upfront payment or sequentially over multiple periods where customers can pay and consume content over time. This sequential release allows customers to learn about the content, drawing from personal experiences (private signals), critics' reviews (public signals), or collective consumer feedback (social signals). First, we demonstrate that when a price-taker content creator sells content at a fixed price, and customers only learn from private signals, sequential release yields lower expected revenue than simultaneous release due to the negative impact of private learning on the sequential release strategy, even though customers' surplus is higher under sequential release. Second, we demonstrate that under some conditions, the content creator can enhance the sequential release's expected revenue through flexible pricing or by encouraging public signals, surpassing a simultaneous release's revenue. Third, we demonstrate that when the prior belief on the attraction level of content is very uncertain, if customers rely solely on private signals, setting a low initial price close to free is optimal for the content creator to encourage learning. However, interestingly, if joint private and public signals influence learning, the content creator will restrict learning by imposing a high initial price. Additionally, we demonstrate that under sequential release, when the market is sufficiently large, the content creator benefits more from social signals than public signals. Our results provide strategic guidance for serial content creators regarding content release and pricing strategies, taking into account the information dissemination and pricing policies of their publishers.
Work in Progress
1. Data-driven Competitive Pricing
Setareh Farajollahzadeh, Ming Hu
Status: work-in-progress
Keywords: Data-driven algorithm, Competitive pricing
2. Market Segmentation: Asymmetric Equilibrium among Symmetric Platforms in Ride-hailing Markets
Setareh Farajollahzadeh, Philipp Afeche, Azarakhsh Malekian
Status: work-in-progress
Keywords: Queuing games, Two-sided markets, Autonomous vehicles
Abstract: We study duopoly competition between symmetric platforms in the ride-hailing market, who serve customers with heterogeneous delay cost. platforms choose their service fee and driver capacity while customers choose the utility-maximizing service. We demonstrate that whenever the number of customers with low delay cost is either low or large, the market reaches a symmetric equilibrium in which both platforms set the same price and capacity, thus serving the same number of customers. However, if the number of customers with low delay cost is moderate, there exists an asymmetric equilibrium in which the market will be segmented between the two types of customers.
Case Study