2021 – Present
This project aims to address the challenges posed by architectural design changes during the design phase and post-occupancy, which can significantly alter how crowds behave during emergencies. Changes in building layout, exit placements, and overall design can lead to confusion, bottlenecks, or even panic during evacuations, increasing risk during disasters. Current tools are inadequate for evaluating the impact of alternative architectural design features or physical alterations to existing structures on crowd flow. The primary objective is to integrate human behavior modeling with architectural design and analysis using digital twins, which are virtual models of physical locations, environments, or systems. This integrated approach will more effectively guide the design and modifications of buildings to influence, promote, and enhance the desired crowd behavior during emergencies, such as active shooter scenarios and bomb threats.
The project utilizes algorithms and computational modeling to forecast crowd behavior in response to architectural designs. It leverages human behavior simulations and 3D digital twins to provide insights into the influence of architectural design on crowd management, security protocols, and environmental analysis. The project seeks to integrate its findings into real-world scenarios by contributing to the center’s vision of the Virtual Sentry Framework. It aims to provide solutions for crowd management, options to adapt physical infrastructure to enhance crowd flow, and tools to inform security directors and patrons decisions during emergencies.
The project improves upon prior SENTRY models to provide advanced capabilities for real-time crowd behavior forecasting. It addresses uncertainties, incorporates new data sources, and scales to larger systems, improving the reliability of simulations and the insights derived from them.
In 2024, the project scanned the Rutgers University Football Stadium using LIDAR, providing a highly accurate and detailed virtual reconstruction that can be used by crowd behavior models to demonstrate crowd flows. It has created a highly parallelized GPU-based crowd simulation that is easily integrated with digital twin platforms.
Researchers collaborate with SENTRY projects in the Layered Security Architectural Design and Simulation Research Area and across all four SENTRY research areas to support the center’s vision of the Virtual Sentry Framework. The project is connected to SENTRY’s Passenger-based Transportation Case Study involving the NJ Transit Hoboken Terminal and SENTRY’s Stadium Security Case Study, which engages Rutgers Football Stadium. These collaborative efforts provide data, testing environments, and real-world applications to validate the research, ensuring the practical relevance of the models and contributing to the comprehensive integration of the Virtual Sentry Framework in real-world applications.
