.png)
For Wendy's, Piloting AI Energy Optimization Starts with Basic Controls Installations
Wendy’s installed new building controls across five restaurants before it could even test AI-driven HVAC optimization.
The move came as part of the company’s effort to reduce absolute scope 1 and 2 greenhouse gas emissions 47% between 2019 and 2030.
Each restaurant uses roughly half its energy for cooking equipment and the other half for building operations, such as HVAC and lighting. That puts building systems squarely in the crosshairs of energy-reduction efforts.
Thomas Grant, Wendy’s Global Energy Manager, partnered with Facil.ai to run a five-restaurant pilot exploring how AI-based advanced supervisory control could reduce energy use.
But before the AI software could run, the restaurants needed basic infrastructure.
The project installed smart thermostats, lighting controls, refrigeration controls, energy meters, and a web-based controls platform across the five sites before Facil.ai’s optimization platform could be deployed.
Wendy’s operates roughly 7,000 restaurants globally, and infrastructure varies widely across locations due to acquisitions and franchise ownership. Some sites lacked functioning building controls entirely, making them unsuitable for evaluating optimization software.
Only after that baseline control layer existed could Facil.ai apply its prescriptive AI platform to optimize HVAC operation.
For Grant, automation also addresses a different operational constraint inside restaurant portfolios. Maintenance teams respond to equipment failures through CMMS work orders, which means inefficient operation often persists when nothing is technically broken.
“If there isn’t a work order created, no action is going to occur to fix the problem,” Grant said. “Our facilities team is not going to fix something without a work order.”
Optimization software targets operational inefficiencies, automatically adjusting control behavior rather than relying on technicians to manually tune systems that are technically still functioning.
The early pilot results were mixed. Kitchens generate large internal heat loads from cooking equipment and ventilation, which means HVAC systems often run near full capacity during peak conditions. When equipment is already operating at maximum output, there is little opportunity for optimization algorithms to modulate setpoints and reduce energy use. Wendy’s has not yet moved to an enterprise rollout.
But the experiment highlighted a practical constraint many portfolios face: energy-optimization software is difficult to evaluate in buildings that lack consistent controls.
For energy managers exploring AI energy optimization, the first step may not be deploying software at all. It may be standardizing the control layer across the portfolio.
Register for the next Nexus Labs event
Sign up for the newsletter to get 5 stories like this per week:
Wendy’s installed new building controls across five restaurants before it could even test AI-driven HVAC optimization.
The move came as part of the company’s effort to reduce absolute scope 1 and 2 greenhouse gas emissions 47% between 2019 and 2030.
Each restaurant uses roughly half its energy for cooking equipment and the other half for building operations, such as HVAC and lighting. That puts building systems squarely in the crosshairs of energy-reduction efforts.
Thomas Grant, Wendy’s Global Energy Manager, partnered with Facil.ai to run a five-restaurant pilot exploring how AI-based advanced supervisory control could reduce energy use.
But before the AI software could run, the restaurants needed basic infrastructure.
The project installed smart thermostats, lighting controls, refrigeration controls, energy meters, and a web-based controls platform across the five sites before Facil.ai’s optimization platform could be deployed.
Wendy’s operates roughly 7,000 restaurants globally, and infrastructure varies widely across locations due to acquisitions and franchise ownership. Some sites lacked functioning building controls entirely, making them unsuitable for evaluating optimization software.
Only after that baseline control layer existed could Facil.ai apply its prescriptive AI platform to optimize HVAC operation.
For Grant, automation also addresses a different operational constraint inside restaurant portfolios. Maintenance teams respond to equipment failures through CMMS work orders, which means inefficient operation often persists when nothing is technically broken.
“If there isn’t a work order created, no action is going to occur to fix the problem,” Grant said. “Our facilities team is not going to fix something without a work order.”
Optimization software targets operational inefficiencies, automatically adjusting control behavior rather than relying on technicians to manually tune systems that are technically still functioning.
The early pilot results were mixed. Kitchens generate large internal heat loads from cooking equipment and ventilation, which means HVAC systems often run near full capacity during peak conditions. When equipment is already operating at maximum output, there is little opportunity for optimization algorithms to modulate setpoints and reduce energy use. Wendy’s has not yet moved to an enterprise rollout.
But the experiment highlighted a practical constraint many portfolios face: energy-optimization software is difficult to evaluate in buildings that lack consistent controls.
For energy managers exploring AI energy optimization, the first step may not be deploying software at all. It may be standardizing the control layer across the portfolio.
Register for the next Nexus Labs event
Sign up for the newsletter to get 5 stories like this per week:
This is a great piece!
I agree.