Projects

Statistical Optimization and Analytics for Community Emergency Management Systems

Project Webpage

This project aims to improve Emergency Response Management (ERM) systems using proactive resource management to minimize response times and maximize response effectiveness. With road accidents accounting for 1.25 million deaths globally and 240 million emergency medical services (EMS) calls in the U.S. each year, there is a critical need for proactive ERM decision support systems. Furthermore, a timely response to these incidents is crucial and life-saving for severe incidents. This is challenging due to dynamic and uncertain environments. The process of effective ERM requires the integration of emergency forecasting models, planning algorithms, incident detection approaches, response policies, and recovery policies. However, current state-of-the-art research has focused primarily on individual aspects of ERM, and the current practice of ERM workflow in the U.S. is reactive, resulting in significant variance in response times.

We look at ERM holistically. We design continuous-time generative models to forecast spatiotemporal incidents. We have also developed efficient and scalable approaches to solve the high-dimensional optimization problem of proactive stationing and dispatch under uncertainty by using Multi-agent Monte Carlo Tree Search (MMCTS).

Selected Publications:

Transit System Optimization

Project Webpage

This project aims to develop algorithms to perform system-wide optimization for fixed line, paratransit, and microtransit services while focusing on the objectives: (1) minimizing energy per passenger per mile, (2) minimizing total energy consumed, and (3) maximizing the percentage of daily trips served by public transit. While it is possible to optimize these decisions separately as prior work has done, integrated optimization can lead to significantly better service (e.g., synchronizing flexible courtesy stops with microtransit dispatch for easy transfer). This is challenging, however, due to uncertainty in the future demand and traffic congestion. We address these challenges using state-of-the-art artificial intelligence, machine learning, and data-driven optimization techniques such as Deep Reinforcement Learning and Monte-Carlo Tree Search.

Selected Publications: