RESEARCH PROJECTS
The Course of My Career
The National Platform for Outage, Welfare, and Energy Resilience(NPOWER)
Summary: Datasets from multiple publicly available federal sources undergo detailed processing and formatting within N-POWER to effectively highlight the effects of electricity outages across the US and their impacts on various demographics. This map-based dashboard is designed to offer statistics at multiple levels: national, state, and county. Users can interactively filter outage causes ranging from severe weather events to transmission/ distribution interruptions to physical attacks. Furthermore, the dashboard presents various demographic information such as racial composition, disability prevalence, insurance coverage distribution, and community resilience metrics. This feature is particularly valuable in identifying areas and demographic groups that are disproportionately affected by different outage reasons, offering crucial insights for targeted responses and policy development.
N-POWER provides a clear, accessible view of energy issues and their effects on communities across the US. It equips users with the data and insights needed for informed decision making. Its comprehensive approach makes it an invaluable resource for anyone looking to understand and address the complexities of energy reliability and societal impact.
December 2022-Present
Team Members: Ziyu Zhao, Hao Lin, Tejas Raj, Jack Weissman, Leo Hu
Queue Routing Strategies to Improve Equitable Housing Coordination
Summary: Although access to housing and support services is a good way to decrease runaway and homeless youth's vulnerability to being exploited, coordination of these services provided to RHY by non-profit and government organizations is neither standardized, nor efficient. This situation often causes decreased, delayed, and inequitable access to these scarce housing resources. Therefore, we aim to increase the housing system efficiency and reduce the barriers that are contributing to inequitable access to housing through simulation modeling and analyses. We simulate a set of crisis and emergency shelters in New York City that are funded by a single governmental organization considering poisson arrival process, generally distributed service and patience times, and multiple server pools. The simulation enables us to evaluate various queue routing strategies that are previously introduced to analytics literature as housing and support service coordination mechanisms and helps us to provide useful managerial insights.
December 2022-Present
Team Members: Dr. Kayse Lee Maass
Priority Thresholds for Equitable Service Access
Summary: One concern within the housing assistance system is equitable access; runaway and homeless youth (RHY) have different characteristics that may make it challenging for them to find a compatible shelter; and that may also increase their vulnerability to exploitation. Therefore, we propose a queuing model to improve equitable access to housing programs for RHY in New York City. The queuing model allows us to: (i) investigate the populations and demographics who are facing access barriers, (ii) project the minimum number of servers (beds) required to provide a certain global service quality level to every youth, and (iii) prioritize RHY who are at more risk at trafficking to promote equity. During this research, we have conducted focus groups with both RHY and service providers to discuss the applicability of the priority policy.
January 2022 - Present
Other Research Team Members: Dr. Kayse Lee Maass
Reducing Vulnerability to Human Trafficking: Improving Access to Housing and Support Services for Runaway and Homeless Youth in New York City
Summary: Recent estimates indicate that there are over 1 million runaway and homeless youth and young adults (RHY) in the United States (US). Exposure to trauma, violence, and substance abuse, coupled with a lack of community support services, puts homeless youth at high risk of being exploited and trafficked. Although access to safe housing and supportive services such as physical and mental healthcare is an effective response to youth's vulnerability towards being trafficked, the number of youth experiencing homelessness exceeds the capacity of available housing resources in most US communities. Therefore, in this study, we developed an integer linear programming model that extends the multiple multidimensional knapsack problem to project the collective capacity required by service providers to adequately meet the needs of RHY. Our RHY-centered approach is an important step toward meeting the actual, rather than presumed, survival needs of vulnerable youth, particularly those at-risk of being trafficked.
January 2021 - May 2023
Other Research Team Members: Geri Dimas, Dr. Kayse Lee Maass, Dr. Andrew Trapp, Dr. Renata Konrad, and Dr. Meredith Dank
Improving Emergency Department Access to Computed Tomography Resources: A Multi-Step Approach
Summary: Hospitals face the challenge of managing demand for limited computed tomography (CT) resources from multiple patient types with different anatomical scan needs, while ensuring timely access. In this study, we have taken a multi-step approach to improve CT access for emergency department (ED) patients at a large academic medical center with six unique CT machines that serve unscheduled emergency, semi-scheduled inpatient, and scheduled outpatient demand. The goal of this study is to develop a schedule that ensures 90% of ED patients to receive service within 80 minutes of their arrival, while also creating capacity to meet demand for inpatients and scheduled patients. Our approach involves a queuing model of ED patients with the possibility of abandonment, and an integer linear programming model that projects the optimal schedule and allocates resources to specific patients. The comprehensive schedule is likely to increase organizational effectiveness when implemented.
January 2021-Present
Other Research Team Members: Dr. Kayse Lee Maass, and Dr. Kalyan Pasupathy
Simulation Model to Evaluate LGBTQ+ RHY Access to Housing Resources
Summary: Runaway and homeless youth and young adults are highly vulnerable to human trafficking - the vulnerability to trafficking increases even more for LGBTQ+ youth. A 10-city study, conducted from 2014 to 2016, found that 24% of LGBTQ+ RHY were victims of sex trafficking, and LGBTQ+ youth account for up to 40% of the RHY population, although only 3-5% of the general youth population identify as LGBTQ+. Therefore, in this study undergraduate research assistants and I present a discrete event simulation model of a crisis-emergency housing program that also provides drop-in services to LGBTQ+ youth in NYC. Our aim is to analyze the current operations and test potential capacity expansion interventions. We use the simulation as a tool to communicate with policymakers, funders, and service providers in NYC.
January 2021- May 2022
Other Research Team Members: Sophia Mantell, Geri Dimas, Dr. Kayse Lee Maass, Dr. Renata Konrad, Dr. Andrew Trapp, and Dr. Meredith Dank
Modeling Hospital-Specific Bed, PPE, and Staff Surge Capacity During COVID-19
Summary: The COVID-19 pandemic has placed significant surge strain on hospital beds, staff, ventilators, personal protective equipment (PPE), and other supplies. This situation raised the need for facility-specific capacity forecasts to make key operational decisions. With a team of interdisciplinary researchers, we present a tool to help health systems estimate and visualize 1-to-30 day ahead hospital-specific demand for medical and ICU beds, ventilators, PPE, medications, and available staff on a rolling basis. With this study, we provide a free online tool that can be used by hospitals around the world to estimate their COVID-19 related needs.
January 2020 - December 2020
Other Research Team Members: Basma Bargal, Shane Yap, Dr. Michael Rosenblatt, and Dr. James Benneyan
Boston Children's Hospital Complex Care Process Redesign
Summary: Children with medical complexity (CMC) are a growing population of children aged 0 to 18 years with multiple chronic and costly medical conditions that often lead to severe functional limitation. One example is spinal fusion, an orthopedic surgery to correct severe curvature of the spine caused by scoliosis. Although effective, these operations pose a significant safety risk for CMC; complication rates exceed 25% for some children. In this study, working with clinicians, physicians, nurses, patients, and patients’ families at Boston Children’s Hospital, we aim to install a reliable preoperative pathway for children undergoing spinal fusion surgery. We develop predictive models, queuing simulations, and perform failure mode and effects analysis. With the implementations, our team was able to decrease the surgical site/instrumentation infections from 15% to 7%, and average length of stay of the patients has been decreased by 0.8 days; based on these values cost of recovery in hospital has decreased approximately $21,000 per patient.
January 2019-December 2020
Other Research Team Members: Dr. Jay Berry, Dr. Sara Singer, Lucia Bastianelli, Elizabeth Casto, Dr. Joanne Cox, Erin Ward, Will Ward, Basma Bargal, Shane Yap, Dr. James Benneyan and more
Improving Efficiency of Trauma Screening Process for Foster Care Children
Summary: On any given day, there are nearly 438,000 children in the United States foster care system and there are multiple trauma screening tools that are used to diagnose the trauma experienced by the child, with varying guidelines for sensitivity and specificity values. In partnership with social workers and sociologists, we propose a multi-step approach including a probability model and a queuing simulation to estimate the effect of using different trauma screening tools on number of under-treated and over-treated children within the system, as well as their expected wait times to receive care. We aim to inform policy makers to decide on the selection of which screening tool to use and what threshold is most suitable to the current system capacity.
January 2019 - January 2020
Other Research Team Members: Basma Bargal, and Dr. James Benneyan
Statistical Monitoring of Queueing Networks
January 2018-December 2018
Summary: Queuing systems are important parts of our daily lives, and to keep their operations at an efficient level they can be monitored using queuing performance metrics, such as average queue lengths and average waiting times. In this study we focus on detecting the change in service rates in open Jackson queuing networks. To do so, we propose using Cumulative Sum (CUSUM) control charts based on likelihood ratios while considering average run length as the performance measure. While this research was conducted purely on simulated data, with the help of our queuing simulations, we were able to come to conclusion that CUSUM charts can effectively detect the change in service rate even if the change in parameters is small.
Team Members: Dr. Devashish Das