INFORMS 2021 Program Book

INFORMS Anaheim 2021

SC20

SC18 CC Room 202B In Person: Energy Equity Flash Session Chair: Ogechi Nwadiaru, University of Massachusetts, Boston, MA, United States 1 - The Energy Equity Gap: Unveiling Hidden Energy Poverty Destenie Nock, Assistant Professor, Carnegie Mellon University, Pittsburgh, PA, 15207-1120, United States, Shuchen Cong, Lucy Yueming Qiu Income-based energy poverty metrics, miss people’s behavior patterns. Our Energy Equity Gap metric measures energy poverty based on user consumption patterns. Specifically, we use a residential electricity consumption dataset from Arizona to determine the temperature at which households turn on their home cooling systems. The Energy Equity Gap reveals that low income households wait 4-7°F longer than high income households to turn on their air conditioning units. In our region of study, the EEG widened between 2015 and 2019. This reveals demand elasticity in low income communities, and may be caused by delayed effects of residential electricity price changes. 2 - Environmental and Economic (In)Justice Considerations When Decarbonizing a Power System Paola Furlanetto, University of Massachussets We investigate the interaction of environmental and equity goals in a network- constrained power system, to identify scenarios where an increase in overall environmental quality may lead to unjust consequences. Using a deterministic unit commitment model, we layer environmental and socioeconomic data over IEEE power flow test cases. We examine the impact of greenhouse gas emissions constraints on air pollution in marginalized neighborhoods and energy burden reflected by locational prices. 3 - Hydrogen as a Transition Alternative for Oil-and Gas- Dependent Countries A Nigeria Case Study Ayoola Fola To mitigate significant damages from our changing climate, CO2-equivalent emissions must reach net-zero by 2050. This can be achieved only by reducing emissions from fossil fuel consumption. Currently, about 70% of this consumption is of oil and gas, making fossil-fuel-dependent countries such as Nigeria, uniquely vulnerable. My work uses a macroeconomic analysis framework with energy systems modeling tools to explore the potential of a hydrogen economy as an alternative to the oil and gas economy for Nigeria in a low-carbon future, with the utilization of currently flared natural gas. The scale of infrastructure required is determined, as well as competition in a global commodity market. 4 - Air Pollution Consequences of Vehicle Electrification in India Tapas Peshin, Stanford University, Stanford, CA, United States Transportation related emissions account for approximately a third of particulate matter pollution in India, and a somewhat higher proportion of nitrogen oxides, another set of compounds harmful to human health. A move towards vehicle electrification can be perceived as sustainable, but the net health and distributional impacts will also be determined by the increase in emissions profile from the coal heavy electric grid. Through this work, we determine that moving towards a sustainable, low carbon and low pollution electricity grid is a requirement to make a widespread transportation electrification case for India aimed at achieving equitable energy goals. 5 - Impact of Electricity Storage Ownership Structures on Community Equity Outcomes Ogechi Vivian Nwadiaru, University of Massachusetts, Amherst, MA, United States The work evaluates different ownership structures on a set of predetermined outcomes. We examine the objective function of decision makers in different storage ownership patterns ranging from community storage to utility owned systems. Specifically we identify reliability, autonomy and cost as a priority for stakeholders.

SC19 CC Room 203A In Person: Data-driven Approaches for Combating Healthcare Challenges General Session Chair: Hrayer Aprahamian, Texas A&M University, College Station, TX, 77840, United States 1 - Optimal Unlabeled Set Partitioning with Application to Risk-based Quarantine Policies Hrayer Aprahamian, Texas A&M University, College Station, TX, 77840, United States, Jiayi Lin, Su Li, Hadi El-Amine We consider the problem of partitioning a set of items into subsets so as to optimize an additive objective. Under an arbitrary objective, this family of problems is known to be an NP-complete combinatorial optimization problem. We study this problem under a broad family of functions characterized by elementary symmetric polynomials. By analyzing a continuous relaxation of the problem, we identify conditions that enable the use of a reformulation technique in which the problem is cast as a more tractable shortest path problem. We demonstrate the impact of the methodology through a novel and timely application of quarantining heterogeneous populations in an optimal manner. Our case study on COVID-19 data reveals significant benefits over conventional measures in terms of both spread mitigation and economic impact, underscoring the importance of data-driven policies. 2 - Risk Reduction and Prevention of Epithelial Ovarian Cancers Michael J. Hintlian, PhD Student, University of Southern California, Los Angeles, CA, United States, Julia L. Higle Epithelial ovarian cancers (EOCs) account for approximately 95% of ovarian cancers and are the leading cause of gynecological cancer deaths. Screening for ovarian cancer has not proven to be cost-effective, but studies identify the fallopian tube epithelium as the origin of most high-grade serous carcinoma (the most common, and lethal, EOC). This presents the possibility for opportunistic and prophylactic risk-reducing procedures (e.g., salpingectomy the removal of the fallopian tubes). We examine the effectiveness of such procedures via model- based analysis. 3 - Heuristic Policies for Spatiotemporal Vaccine Allocation Based on Motivated by the COVID-19 pandemic, we study how a government agency may dynamically allocate vaccines from a limited stockpile to different jurisdictions. A generalized SEIR model with behavioral feedback is proposed. Behavioral feedback depends on time varying local transmission rates, which can be extracted from reported death counts. The SEIR model is used to evaluate a variety of implementable allocation policies (e.g., pro-rata policy, allocation proportional to infection rate, allocation proportional to number of susceptible individuals or allocation focused on regions with highest infection rates) in terms of their efficiency and fairness. SC20 CC Room 203B In Person: Stochastic Optimization in Healthcare General Session Chair: Behshad Lahijanian, University of Florida, Gainesville, FL, 32611-6595, United States 1 - Survival Optimization Problems for Cardiac Arrests Dmitry Anokhin, George Washington University, Washington, DC, 22209-3210, United States, Miguel Lejeune We propose new survival optimization models (SOMs) that implement the idea of survival function in emergency healthcare. The key feature of the SOM models is the incorporation of the survival function, which is an isotone function of the overall response time. The response time is defined as the sum of the travel and waiting times. We model the waiting time as an endogenous source of uncertainty in order to capture the impact of decisions on the waiting time. a Compartmental Model with Behavioral Feedback Julius Barth, University of Texas at Austin, Austin, TX, United States, Diwakar Gupta

21