AI for Public Safety

Exploring AI Use Cases to Support Firefighter Safety and Operations.

OVERVIEW

This project aimed to explore how AI-powered tools can enhance public safety, focusing on firefighters during active scenes.

My role

UX Researcher

Duration

8 weeks

Team

5 members

Tools

Miro, Google forms

Client

First responder network authority

THE CHALLENGE

Firefighters face numerous dangers during operations, with limited access to real-time, actionable information.

THE CHALLENGE

How can AI tools provide data-driven insights to improve safety and support firefighters in the field?

PROJECT GOALS

1

3

Identify the key difficulties firefighters face during active scenes.

2

Understand what tools are currently used by firefighters and explore how AI can enhance them.

Evaluate the benefits and limitations of AI products already in use.

4

Explore firefighters' hesitation around adopting AI technologies.

RESEARCH METHODOLOGY

Desk Research

Phase 1

Interviews

Phase 2

Field Observations

Phase 3

DESK RESEARCH

What tools and technologies are currently available to support firefighters, and how can AI improve their effectiveness while addressing ethical concerns?

We began our research by exploring a variety of sources, including online databases, articles, and research papers to expand our understanding of the tools and technologies used by firefighters today.

INSIGHTS

Current Tools & Limitations

Radios

Essential for communication on an active scene but face interference in large structures.

Thermal Imaging Cameras

Crucial for visibility in smoke-filled environments, but can be difficult to operate under pressure.

Navigation Systems

Helps firefighters navigate during emergencies but lack real-time adaptability in fast-changing environments.

Personal Protective Equipment (PPE)

PPE can be heavy and cumbersome, limiting movement and reaction time.

Potential for AI solutions

AI could provide real-time mapping with drones, helping firefighters navigate difficult terrains and avoid hazards.

AI-integrated wearables could monitor firefighters’ health, send alerts if there’s a danger of overheating or exhaustion.

AI-enhanced communication systems could offer uninterrupted data sharing, even in low-signal areas.

WHAT'S NEXT?

This phase laid the groundwork for our next step: interviewing firefighters to get their firsthand feedback on the tools they use, their thoughts on AI, and the specific challenges they encounter in their day-to-day operations.

INTERVIEWS

What challenges do firefighters face on the field, and how do they perceive AI technology as a solution to these challenges?

After gathering insights from secondary research, we conducted semi-structured interviews with firefighters to better understand their day-to-day challenges and thoughts on adopting AI tools. These interviews helped validate our previous findings while providing new, firsthand insights into the practical needs of firefighters.

6 firefighters

We conducted interviews with 6 firefighters from different fire departments to gather insights on their day-to-day challenges and perceptions of AI in firefighting.

Potential for AI solutions

Perception of AI

Current Technology Challenges

Firefighters mentioned frequent issues with radios in large or complex structures.

Many tools, like thermal imaging cameras, are difficult to operate in high-stress situations.

Monitoring vital signs during emergencies is tough, especially when firefighters push themselves to the limit.

Firefighters expressed hesitation about relying on AI citing concerns about trust.

Reluctance toward full AI systems & interest in using AI to improve existing tools, like thermal cameras & communications.

Any AI-based tool would need to be intuitive and easy to use with minimal training.

Thermal cameras are critical but hindered by response times and difficult interfaces.

Health monitoring with real-time vitals tracking is lacking; AI wearables offer a potential solution.

Firefighters need concise, actionable data rather than excess information, reducing data overload.

FIELD OBSERVATIONS

What challenges do firefighters face in real-time operations, and how can AI provide practical, real-world improvements?

Firestation in Greenwoord, Indiana

We closely observed firefighters during training exercises to understand how they use their current tools and where AI could improve their operations.

Thermal Cameras in Real-World Use

We observed firefighters using thermal cameras to navigate through dense smoke, their slow response time meant that firefighters had to pause, causing delays in decision-making, and interpreting the camera’s output was challenging in fast-moving scenarios.

Physical Exertion Without Monitoring

The physical demands of firefighting were evident during our observations, they pushed themselves to the limit, with no real-time system in place to monitor their health (e.g., heart rate, temperature). The lack of health monitoring increased the risk of exhaustion or heat stroke.

Communication Gaps in Large Structures

Radios were frequently used, but signal interference occurred in multi-story and large buildings. This forced teams to repeat instructions or stop moving to ensure they heard correctly, leading to delays.

AI USE CASES

After conducting interviews and observations, we realized that firefighters face significant usability issues with their current tools. The feedback revealed that while they are open to AI, the technology must be intuitive, easy to use, and seamlessly integrated into existing workflows. From there, we began developing AI-driven solutions focused on real-time responsiveness, health tracking, and communication enhancement.

How can AI make real-time decision-making faster and more efficient?

AI-Powered Communication System (AUDREY)

Firefighters need to navigate complex fire scenes quickly, but visibility is often limited, and real-time data is scarce.

AUDREY integrates sensor data, video feeds, and environmental data, ensuring seamless information sharing and communication.

Benefit: Faster, uninterrupted communication that improves coordination during emergencies.

What systems can improve firefighter safety, particularly in health and environmental monitoring?

Fire Neural Network (FNN) for Predicting Fires

Predicting fire behavior, especially in forested areas, is difficult.

The Fire Neural Network (FNN) analyzes lightning strike data to predict areas at high risk for fires. It also uses drones to confirm ignitions and provides real-time alerts.

Benefit: Real-time fire behavior predictions allow firefighters to deploy resources where needed, preventing uncontrolled spread.

FireScout AI for Wildfire Detection

Wildfires spread unpredictably, making it hard for firefighters to respond quickly.

FireScout AI uses satellite imagery and thermal data to detect wildfires in their early stages, providing early warnings and resource recommendations.

Benefit: Early detection and timely resource deployment reduce the impact of wildfires and improve firefighter safety.

SOME OTHER USE CASES

LiDAR Drones for Real-Time Fire Mapping

Navigating complex fire scenes with limited visibility is a challenge.

AI-powered LiDAR drones generate real-time 3D maps, helping firefighters and commanders assess the terrain, locate hazards, and plan resources.

Benefit: Faster, more informed decision-making in chaotic environments.

AI-Integrated Wearable Health Monitors

Firefighters have no real-time tracking of their vitals, risking overexertion.

Wearable AI devices track heart rate, body temperature, and hydration, sending alerts when vitals are in danger zones.

Benefit: Firefighters can be pulled out before reaching critical fatigue, reducing health risks during high-stress situations.

NEXT STEPS

1

User Testing: Gather feedback from firefighters to further refine these AI tools.

2

Address Legal & Privacy Concerns: Work with experts to ensure AI solutions comply with safety and privacy standards.

3

Competitive Analysis: Conduct further research to compare our proposed AI solutions with existing technologies.