Can AI Run Your Fleet?
Despite growing enthusiasm for artificial intelligence, the most important fleet decisions continue to rely on experience, context, and critical thinking.
I had a conversation recently with a LinkedIn contributor who is an influencer on fleet management issues. The subject was the use of artificial intelligence (AI) in making fleet management decisions. In a post, the contributor said that AI is becoming indispensable for managing a fleet.
My view is simple: AI is no substitute for knowledge built through education and experience.
It can be helpful for tasks such as creating presentations or drafting articles like this one. However, decisions in many operational settings—including fleet management—still depend on practical know-how and sound judgment.
Ask Chat GPT
Here is an example. I asked ChatGPT to tell me how to reduce carbon emissions in a fleet of light, medium and heavy-duty trucks. This is part of the response:
Here are the main levers companies use today: Battery-electric trucks: best for many light-duty and some medium-duty urban routes with predictable daily mileage and depot charging.
Hybrid vehicles: useful for stop-and-go delivery operations where full electrification isn’t practical yet.
Renewable diesel (HVO): Can often reduce lifecycle emissions substantially while using existing diesel engines and fuelling infrastructure.
Natural gas / RNG: sometimes effective for heavy-duty regional hauling, especially where renewable natural gas is available.
Hydrogen fuel cell trucks: emerging option for long-haul heavy-duty applications, though infrastructure is still limited.
I agree that these are all possible solutions, but I know there are supply constraints from having looked at them for many years. The availability of renewable fuels may be good in some parts of the country and non-existent in others.
Hydrogen fuel cell trucks are still not available, after twenty years of talking about how they will solve the problem of carbon emissions in long-haul trucking. Hydrogen for buses is also doubtful, after numerous failures in municipal transit.
Obtaining hydrogen fuel is still impossible unless a fleet operator is willing to make it themselves. That requires an enormous investment in infrastructure and technical knowledge.
There is also the issue of higher cost that is a barrier for many fleets that government incentives can help alleviate, but not overcome. The business case for a truck running on natural gas may be there if only carbon emissions are considered, but if the cost of converting the truck engine and then maintaining it is considered, the business case usually fails.
“AI is no substitute for knowledge.”
Experience still matters
I think fleet operations still require real know-how. In the same way a doctor can interpret blood test results by understanding what the numbers mean, experienced fleet professionals can look at information in context and make informed decisions.
AI can help people get started, especially if they lack that background. But when a decision calls for critical thinking, practical judgment, and operational experience, AI is not a dependable substitute.
