Edmonton leveraging AI to make safety code inspections more efficient

Using risk-based predictive modeling to score pass rates for non-mandatory inspections, the City has been able to devote more attention to higher-risk projects while shortening timelines for builders.

Artificial Intelligence (AI)Predictive AnalyticsDataPublic Safety
Edmonton safety inspection AI wins Smart 50 Award -
EDMONTON, ALTA. - The City of Edmonton is using the future to study the past and speed up present building. Edmonton has been awarded a Smart 50 award in the digital transformation category by Smart Cities Connect for an artificial intelligence (AI) solu
Journal Of Commerce
Smart City Awards & Recognition :: City of Edmonton
Edmonton’s Smart City efforts have garnered attention from around the world.
  • The City of Edmonton uses nearly a decade of data to deploy AI model to increase inspection efficiency and effectiveness
  • Since October 2019, the predictive model has reduced the number of eligible inspections by 37%
  • Inspections deemed low risk are passed automatically, eliminating unnecessary delays in builder timelines
  • City inspectors are able to focus on higher risk and more complicated inspections, which pose greater threat to safety
Project Summary
Using data collected from over 600,000 safety inspections completed over the course of a decade, the City of Edmonton has trained an AI model to predict whether single-family home builders will pass low-risk inspections.
And they’ve put this data to work by enabling the City to automatically pass inspections for builders with a good track record and for those inspections posing limited risk to public (homeowner) safety. This model has reduced builder delays and allowed the City to focus resources on higher risk inspections.
In fact, the use of this AI model and the new inspection process in Edmonton has resulted in a 37% decrease in eligible inspections – a staggering efficiency pickup – and an economic boost from faster development with fewer roadblocks.
With finite resources and variable risk for building projects, this shift to more data-driven resource allocation seems logical, but has proven tougher to implement across agencies and departments for various reasons such as data silos and staffing limitations.
While some US-based municipalities have already begun to work with predictive modeling (such as Pittsburgh, PA and Atlanta, GA), Edmonton is the first city in Canada to use AI in a risk based predictive model to make Safety Codes inspections more efficient.
It's also earned them a global Smart 50 Award for digital transformation in 2020.

Towards better decisioning-making

The City is bringing in new data constantly as builders continue to apply for permits and new scenarios are run through the model. The next step for Edmonton is to expand the model outside single-family residences and increase the scope of impact.
And while the buzz around smarter use of data continues to grow, Edmonton mayor Don Iveson stresses the importance of creating policies around this data.
“I truly believe the next frontier for governments is how we can use data and AI to make better evidence-based decisions. To know that we’re leading the way in this rapidly-evolving field while making things easier for businesses is very encouraging,” Iveson said.
🚫You must be signed in as a Government user to join the discussion.

Lindsay Pica



In Progress
🏅 Awards
Smart 50 Awards - Digital Transformation
Smart City Edmonton
Related Stories
Boulder Home Preparedness Assessment Pilot Program
Boulder, CO helped homeowners enhance their personal resilience to natural disasters and other stresses through an innovative pilot program that included rebates and free consultations.
The Atlanta Fire and Rescue Department is using the open-source framework Firebird to predict fire risk and prioritize fire inspections.
Show more