Using AI to improve fire safety for First Nations communities

Fire safety asset detection exceeds 90% accuracy 

Fire related deaths in Canada’s First Nations communities are 10.2X higher compared to other communities. At the same time, 90% of their fire incidents are associated with inadequate or non-functional fire safety assets. Compounding the problem is the shortage of attainable and affordable fire insurance for First Nations communities.¹

To help save lives, improve household fire safety and increase access to affordable fire insurance for First Nations communities, Mustimuhw Information Solutions (Mustimuhw) partnered with CGI to develop a solution demonstrating the responsible use of artificial intelligence (AI) and advanced, non-intrusive technology in a two-phased pilot program to achieve these goals. 

Phase 1: Create fire safety digital twins

The first phase used AI and machine learning (ML) techniques to build digital representations of buildings, called digital twins, and risk models to identify potential threats based on fire protections in place and the ability to detect and respond to fire events. AI models were then used to detect, identify and recognize fire safety elements within these buildings.  

Phase 2: Monitoring and management of the digital twins 

The second phase provides real-time digital monitoring and management of the buildings’ environmental conditions. This includes: 

• Deploying IoT sensors within the structures to remotely monitor their functionality and environmental variables, such as sudden temperature fluctuations, air quality, dampness and mold in real-time. 
• Using scalable cloud services to build complex rules to establish thresholds and manage the sensors to monitor, alert, and support fire responses appropriately.
• Harnessing AI to create a Digital Twin to Manage and Alert Operations Center that provide First Nations building managers with interactive visualizations and reports through the Mustimuhw Indigenous Digital Health Ecosystem. These insights support fire safety measures, alerts to potential issues, and deliver aggregated results back to the community.