Wednesday, February 19 from 12:00pm – 1:00pm ET
SENTRY Soft Target Security Series: Real-Time Threat Detection & Mitigation

By invitation only.
Contact t.lam@northeastern.edu with questions.
David Castañón, Boston University
SENTRY has a research area titled Real Time Threat Detection and Mitigation, which is exploring potential decision support systems for command centers. The area has four major projects, working on topics from video analytics, multisensor information collection and fusion, behavioral models of humans in threat environments, and recommendation systems for threat mitigation. This webinar will overview selected results from two of the projects in the area.
Humanizing Agents in Agent Based Models for School Shooting Simulations
Richard John, University of Southern California
With the ever-growing threat of school shootings, modeling these tragedies is crucial to mitigate or reduce casualties in future events. However, agent-based models traditionally instruct agents to act based on theoretical behavior rather than actual human behavior. We present the results of 81,000 simulations of a school shooting where agent behavior is modeled after actual human behavior from a similar virtual scenario. Agents’ reaction time and movement speed are drawn from probability distributions based on human data. The pathways assigned to agents are based on the human behavior exhibited in response to three social influence conditions: where the non-player characters all ran, all hid, or both ran and hid. Additionally, we manipulate law enforcement dispatch time, shooter accuracy, and magazine capacity. Results show mixed agent behavior and lower dispatch times had the largest influence on casualties. The methodology demonstrates the power of empirically defining agent behavior in ABMs.
Learning Enabled Real-Time Strategies for Safely Avoiding an Active Shooter in Unknown Environments
Mario Sznaier, Northeastern University
Milad Siami, Northeastern University
This presentation will cover current SENTRY activities towards developing a real-time decision support system for safe evacuation in active shooter scenarios. We will briefly cover a new predator-swarm-guided particle (PSG) model of a crowd being guided to safety in the presence of a threat (for example children being guided by a teacher or sports event attendees being guided by ushers). We will also illustrate the use of state-of-the-art machine learning tools to generate real-time strategies to steer this guided particle to safety in partially known scenarios. A salient feature of our approach is that it allows for “on-the-fly” re-planning as new obstacles or threats appear in the scene.
Advanced Real-Time Evacuation Routing for Active Shooter Scenario
Subhadeep Chakraborty, University of Tennessee
This presentation will discuss some new methods for improving decision-making during active shooter incidents. Using a combination of AI tools and simulations based on real school layouts, we will talk about a system to identify the best evacuation routes while considering factors like safe rooms and controlled door locking. Some of the key breakthroughs include a strategy to contain threats with equitable risk distribution and development of a digital simulation platform to test and improve our approach, enhancing safety and emergency response plans.
David Castañón | Boston University
Professor David Castañón received his Ph.D. degree in Applied Mathematics from the Massachusetts Institute of Technology in 1976. At Boston University, he served as Department Chair, and as co-director of the Center for Information and Systems Engineering. He served as a Thrust Leader in the Department of Homeland Security’s Emeritus Center of Excellence, ALERT and serves as a Thrust Leader for Soft-target Engineering to Neutralize the Threat Reality (SENTRY).
Richard John | University of Southern California
Richard John is Professor of Psychology at the University of Southern California. His research focuses on normative and descriptive models of human judgment and decision making and methodological issues in the application of decision analysis and probabilistic risk analysis (PRA). Richard has consulted on a number of large projects involving expert elicitation, including analysis of nuclear power plant risks (NUREG 1150) and analysis of cost and schedule risk for tritium supply alternatives.
Mario Sznaier |Northeastern University
Mario Sznaier is currently the Dennis Picard Chaired Professor at the Electrical and Computer Engineering Department, Northeastern University, Boston. Prior to joining Northeast- ern University, Dr. Sznaier was a Professor of Electrical Engineering at the Pennsylvania State University and also held visiting positions at the California Institute of Technology. His research interest include safe learning enabled control of cyberphysical systems , applications of dynamical systems theory to Machine Learning, robust identification and control of hybrid systems, robust optimization, and dynamical vision.
Subhadeep Chakraborty | University of Tennessee, Knoxville
Subhadeep Chakraborty is an Associate Professor for the Mechanical, Aerospace and Biomedical Engineering Department at University of Tennessee, Knoxville.