INFORMS 2021 Program Book

INFORMS Anaheim 2021

TD37

3 - An Electric Vehicle Charging Station Access Equilibrium Model with M/D/C Queueing Bingqing Liu, New York University, New York, NY, United States, Theodoros Pantelidis, Stephanie Tam, Joseph Y. J. Chow The use of electric vehicle (EV) fleets is highly dependent on charging infrastructure. Three contributions are made. First, we propose an EV-to-charging station equilibrium assignment model with a nonlinear objective to evaluate charging station configurations. Queueing is modeled as M/D/C queue. Second, to address the non-differentiability, we propose a solution algorithm based on the Method of Successive Averages. Third, the model is calibrated to the NYC DCAS fleet and charging station configuration as of July 8, 2020, and applied to evaluate hypothetical charging station investments based on two alternative strategies. Results are promising for a policy based on high utilization ratio. TD37 CC Room 210C In Person: Systems Engineering in Support of National Security General Session Chair: Isabella Sanders, United States 1 - Military Readiness Modeling: An Actionable Data Framework Connor McLemore, CANA Advisors, Reno, NV, 89509-2342, United States Although the purpose of the Department of Defense (DoD) is accepted broadly to be “to provide ready and sustainable military forces to protect the nation’s vital interests,” the meaning of that statement is largely reliant upon the definition of the word “ready.” Yet it is generally unclear what it means to be ready. Ready for what? How ready? By when? To address this problem, we recommend the DoD adopt a simple, interpretable, and actionable data framework using stochastic scenario libraries. If implemented by the military, such a framework could allow mathematically coherent readiness estimates to better communicate “how ready for what” combinations of military assets are. Additional details can be found in MOR Journal 2021 Vol. 26, #1, “Military Readiness Modeling: Changing the Question from ‘Ready or Not?’ to ‘How Ready for What? 2 - Hybrid Supplier Risk Assessment and Identification Methodology for the Defense Industry Isabella Sanders, United States This paper aims to present a supplier risk identification and assessment framework that rigorously examines the financial outlook of firms and their respective plants, focusing on disruption and disaster risk factors. This hybrid data-driven risk analysis methodology is practical to implement and can be used proactively by firms to improve the stability of their supplier base through risk assessment and reduction. Chair: Dmitrii Sumkin, INSEAD, Singapore, 138676, Singapore 1 - Pricing in Service Platforms: Who Should Set the Prices? Tolga Dizdarer, Wharton School of Business, Philadelphia, PA, 19104-3615, United States, Gerard P. Cachon, Gerry Tsoukalas Motivated by emergence of blockchain-based decentralized service platforms and Uber’s recent driver-pricing practice in California, we investigate how a platform with large supply should set its fares when service providers are heterogeneous in costs. We use a stylized model to compare two prevalent methods in practice: platform-pricing, where the platform sets the prices for all servers, and server- pricing, where prices are defined by the competitive equilibrium of server decisions. We, then, compare these methods to an optimal contract. 2 - On the Financial Inclusion and Sustainability Benefits of Blockchain Adoption in Agriculture Basak Kalkanci, Georgia Institute of Technology, Atlanta, GA, 30308-1149, United States, Saed Alizamir, Foad Iravani An emerging financial innovation enabled by the Blockchain technology in agricultural supply chains is the capability to “tip the farmers.” This innovation empowers socially-conscious customers to identify the individual farmers of their sustainably-sourced products and reward them by sending them direct payments. We examine the implications of this new capability on farmers’ and consumers’ welfare, and agricultural firm profits. We find that tipping capability can make farmers and consumers worse off in expectation, or may increase income disparity among farmers. TD38 CC Room 210D In Person: Blockchain in OM General Session

3 - Optimal Cash Management with Payables Finance Xiaoyue Yan, Cornell University, Ithaca, NY, United States, Li Chen, Xiaobo Ding Payables finance provides a supplier with the option to receive a buyer’s payables early while allowing the buyer to extend its payment due date. Its recent adoption of blockchain technology has made the process more efficient and secure. In this paper, we study the supplier’s optimal cash management policy under such a “frictionless” payables finance arrangement, based on which we quantify the value of payables finance to the supplier and also determine the equilibrium payment term extension for the buyer. Our work extends the classic cash management models to allow all interest gains and costs to accrue together with the cash balance. Our analysis reveals that the optimal cash policy has a cash balance-dependent (L, M, U) structure. We show that it is the cash liquidity enabled by payables finance to hedge cash flow uncertainty that generates value to the supplier. 4 - Designing the Supply of Digital Collectibles Markets Using NFTS Pavel Kireyev, INSEAD, 12 Boulevard Richard Lenoir, Paris, 75011, France, Dmitrii Sumkin, Serguei Netessine Many organizations, such as Formula One and the NBA, issue digital collectibles traded on blockchain-based marketplaces to attract new customers. For some products, traders can create a limited number of new assets from the assets they already have. It affects reselling price and market liquidity and ambiguously impacts the expected gain from reselling. Given the recent growth of some markets and the rapid shutdowns of others, we find the optimal design of the market supply, where there is enough gain from trade and enough liquidity on the market. We estimate a structural model of trade of digital assets and infer how varying transaction costs, the possibility to create new collectibles from existing ones, and the production rate of the new collectibles affect the competition in the market, customer’s surplus from trade, and the firm revenue. TD39 CC Room 211A In Person: Models to Inform Health Policy and Disease Control General Session Chair: Sze-chuan Suen, University of Southern California, Los Angeles, CA, 90089-0193, United States 1 - Inverse Fractionation In Radiotherapy Archis Ghate, University of Washington, Seattle, WA, 98155-5917, United States The objective in cancer radiotherapy is to maximize tumor-kill while limiting toxic effects of radiation dose on nearby organs-at-risk. Given a fixed number of treatment sessions, planners thus face the problem of finding a dosing sequence that achieves this goal. This is called the fractionation problem. Mathematical formulations utilize the linear-quadratic (LQ) framework to characterize radiation dose-response of tumors and organs-at-risk. The optimal dosing plan in this forward problem depends on the parameters of the LQ model. Unfortunately, these parameters are difficult to estimate. Current debates thus focus on the following question: what parameter values will make specific dosing plans effective? I will present an inverse optimization approach to answer this question. 2 - Optimizing Social Distancing Policies: A Dynamic Programming Approach for Coupled High and Low Risk Populations Peng Dai, Industrial and System Engineering University of Southern California, Los Angeles, CA, United States, Sze-chuan Suen Reducing transmission may be an effective way to control disease, but it is not clear when and who needs to social distance in a pandemic scenario, particularly when policies are allowed to change dramatically over time and population subgroup. We construct a Markov decision process model and build an age- stratified SEIR model to identify the optimal policy to maximize social utility for COVID-19. We compare our optimal policies across several regimes and assess differences in resultant utility, number of infections, and deaths over the time horizon. Our results show that the additional flexibility of policies varying over time and population could generate substantial utility gain.

136