Fleet operators in both the public and private sectors are moving artificial intelligence from pilot programs to daily operations with the goal of measurable improvements in safety, uptime, and customer service.
Guardian Bus Company first introduced AI to its operations in response to rising insurance costs and preventable accidents. Executive Vice President Corey Muirhead said the company introduced AI-enabled dashcams to address driver behavior and reduce risk in real time.
“We use AI on two fronts: safety and customer service,” Muirhead said. “When it comes to safety, we use AI dash cams to monitor driver behavior. These AI dash cams were installed to reduce risk and change driver behavior in real time. Our insurance was skyrocketing and we were suffering from a history of losses from avoidable accidents.”
simple facts
guardian bus company
- Vehicle focus: student transportation
- AI application: AI drive recorder for driver behavior monitoring. A phone agent for parents using AI. Route optimization and sick call support are under development
- Safety Process: Integrates AI dashcam data with GPS and telematics to generate weekly and monthly driver safety scorecards
- Reported results: Improved driver performance, reduced frontal collisions and lane departure violations, reduced cell phone and headphone use, and long-term insurance premium savings.
- Organizational hurdles: driver and union agreement on surveillance technology
The impact was both immediate and long-term, Muirhead said.
“Driver behavior and performance has improved significantly. They don’t use headphones or cell phones, and there are fewer frontal collisions and lane departure violations,” he said. “Over the long term, we’ve seen significant reductions in insurance premiums. The safety culture of the company has completely changed.”
Guardian combines AI dashcam data with GPS and telematics input to generate a driver safety scorecard. Recordable violations are reviewed on the safety dashboard, which generates weekly and monthly reports that the safety team analyzes to identify trends. Drivers experiencing problems will be called for retraining and on-road instruction will be provided if necessary.
“AI data is good, but the more you use the data, the better it gets,” Muirhead says. “Automation starts the process, but human oversight still dominates the process.”
Beyond safety, Guardian leverages AI to manage peak call volumes for school drop-offs and pick-ups. At the beginning of a new school year or summer school, call centers are flooded with inquiries from parents. The company currently uses an AI phone agent that recognizes student profiles, cross-references route numbers and GPS data, and provides estimated arrival times.
“Parents have a right to know where their children are, and if our AI phone system can help, that’s our ROI,” Muirhead said, adding that the system is designed to complement, not replace, active agents.
Looking ahead, Guardian believes route optimization and sick call handling are the most pressing opportunities to further drive the value of AI. The company is building driver profiles so that when a driver calls out, the system can determine which drivers are available to cover the route and push instructions.
“This optimization simplification will be a godsend for transit operators,” Muirhead said, pointing to the ongoing driver shortage and the complexity of daily dispatch decisions.
On the local government side, the City of Long Beach, California, expanded AI across its “1,600 vehicle fleet” and integrated Pitstop AI with Geotab telematics to move from preventive to predictive maintenance.
“We wanted to move from preventive maintenance to predictive maintenance,” said Eric Winterset, Fleet Services Manager for the City of Long Beach.
After a two-year pilot mode, the city rolled out the platform to its entire fleet. Winterset said key performance indicators show that unplanned breakdowns have decreased by 20%, towing costs have decreased and fleet availability has increased to 93% compared to the previous average of 90%.
To ensure AI insights drive action, the city equipped supervisors and maintenance teams with customized dashboards that display the status of the vehicles they monitor. Critical codes now trigger email and text notifications, and technicians can review health reports during service visits to address non-critical issues before they escalate.
simple facts
City of Long Beach Fleet Services
- Number of vehicles: 1,600 vehicles/equipment
- AI platform: Pitstop AI integrated with Geotab telematics
- Main objective: Transition from preventive maintenance to predictive maintenance
- Reported results: 20% reduction in unplanned breakdowns, reduced towing costs, and increased vehicle availability from 90% to 93%.
- Next steps: AI-powered part identification, inventory checking, vendor ordering, and proactive component replacement based on trends
“This is just another tool in the toolbox for our staff, and we explained it that way,” Winterset said. “If it helps them be more productive and make diagnostics less frustrating, and knowing they’re working on a unit in the shop and on the side of the road before it breaks down, that all appealed to them.”
Initial doubts from the staff were a hurdle. Winterset recalled an early conversation with a mechanic preparing for a trip about a no-start problem. The health report predicted that the battery voltage would drop and that starting would be possible within three to five days.
“The predictions in that report were four days old,” Winterset said. “He was surprised by the accuracy of the data, but then he started believing in the product.”
Long Beach’s next step is deeper integration between predictive diagnostics and parts management. Winterset envisions a platform that allows customers to identify the exact part number they need, check inventory, and initiate orders with contract vendors before the vehicle is brought into the store. He also expects AI-powered trend analysis to support similar sector-wide proactive campaigns, such as replacing components across model years before widespread failures occur.
“I have been involved in AI efforts for fleets for over 10 years, and while AI has grown tremendously, I believe we have only scratched the surface of its capabilities,” Winterset said.
In both vehicles, the message is consistent. Rather than replacing humans, AI will reshape the way safety, maintenance, and dispatch decisions are made, delivering measurable operational results.
