�����Թ�

A new �����Թ� Board of Regents' Research Grant aims to advance the future of disaster response through an artificial-intelligence enhanced emergency management system. In partnership with the �����Թ� Department of Emergency and Military Affairs and �����Թ� State University, this initiative harnesses AI to enhance disaster planning and response across the state. From predicting wildfires and floods to identifying emerging threats that could overwhelm critical infrastructure, the AI-enabled system will help emergency planners anticipate and mitigate crises before they unfold.

Leading the effort is Maj. Gen. Kerry Muehlenbeck, DEMA director and adjutant general of �����Թ�, and her team of analysts, who emphasize that AI is a tool to support — rather than replace — human decision making in safeguarding �����Թ� communities. The Board of Regents approved funding for the initiative on Thursday, reinforcing its commitment to innovation that strengthens public safety.

“Our public universities are leaders in advancing the development of new technologies central to our state and nation’s public safety, security and stability,” said ABOR Chair Cecilia Mata. “This groundbreaking AI collaboration between DEMA, ABOR and ASU is a shining example of our commitment to deliver transformative research and innovation that addresses pressing safety needs today and drives positive change into the future.”

The AI-enabled platform will enhance DEMA's emergency management capabilities in several key areas, including data analysis, predictive modeling and course of action development. By leveraging AI's data processing power, the system will enable emergency managers to make quicker and more informed decisions, ultimately saving lives, property, infrastructure and taxpayer dollars.

“AI for emergency management preparedness is absolutely exciting. The Regents’ Grants bring that opportunity to blend the brainiacs at the university with the functional practitioners down on the ground,” said Muehlenbeck. “Disaster preparedness and response requires data gathering and an analytical assessment of that data. Those things demand a lot of brain power. To be able to do those things at the speed of a machine is going to just knock our capabilities out of the ballpark.”

The new system will be developed by a team of ASU faculty and student researchers, and ASURE, the university’s specialized research arm focused on impact for national security innovation. The AI system will go beyond traditional disaster forecasting by analyzing potential second- and third-order consequences. For example, the model could predict not only a wildfire's path but also resulting traffic jams and medical facility overloads. It also will be able to explain why it offered specific solutions.

This “explainable AI” approach will provide policymakers with a transparent rationale for recommended courses of action. This approach will enable DEMA to make more informed decisions, taking into account the potential cascading effects of disasters.

“This is about making true impact for the state and helping make �����Թ� a leader in emergency management,” said Paulo Shakarian, Ph.D., the project’s principal investigator and a tenured ASU associate professor in the School of Computing and AI at the Fulton Schools of Engineering. “When you're dealing with a disaster situation, the ability to respond is really tied to how well prepared you are. The collaboration with DEMA will help us understand how this system, that we're going to initially field in �����Թ�, could potentially have broad impact for other states as well.”

Shakarian’s co-principal investigators are Dr. WenWen Li, Dr. Hua Wei and Dr. YooJung Choi. Ted Senator, who has over 40 years of working with AI, including its earliest iterations, will lead ASURE’s specialized confidentiality role in the collaboration.

This team will work directly with Muehlenbeck and DEMA’s division of emergency management director Gabe Lavine.

The DEMA-ASU partnership has a three-year timeline built on three distinct phases. Year one will focus on emergency information analytics and root cause analysis. Year two will concentrate on disaster prediction, including second- and third-order effects. The final year will prioritize developing optimal, explainable courses of action. To accelerate the development process, ASU’s team will provide major software releases to DEMA in six-month intervals and regular releases every three months.

“This is a unique collaboration, with one ASU program supporting another ASU program, supporting a state agency, supporting the community,” said Lavine. “With the funding that's backed by the �����Թ� Board of Regents to do those things, we're reinvesting in the state's future and reinvesting in government industries that that not only help �����Թ�ns, but help our community of practice at large.”