Podcasts

Adiona Tech’s Richard Savoie on Your AI Injection: why drivers ignore "perfect" routes

November 19, 2026

Your AI Injection · 4 February 2026 · hosted by Deep Dhillon

Our CEO Richard Savoie joined Deep Dhillon on Your AI Injection, the podcast from AI consultancy Xyonix. They talked about the drivers behind last-mile delivery, and why route plans that look perfect on paper often fail on the road.

Your AI Injection podcast episode artwork

Listen: Episode page · YouTube · Apple Podcasts · Spotify

Why delivery drivers reject optimized routes

Classic routing algorithms produce what Richard calls “flower petal” routes. Every van loops out from the depot and back, crossing over everyone else's run. The math says that's the most efficient plan. Drivers look at it and think it's wrong. "Humans have this massive psychological rejection of that being efficient," Richard said.

Plenty of routing projects have failed this way. The CFO signs off on the savings, then the drivers refuse to run the routes. With 450,000 manufacturing and transport jobs unfilled in the US, Richard warned that drivers who get fed up will leave for a competitor, and ask for 10 to 20% more money if they come back.

Adiona blends the math with what drivers prefer, plus traffic and weather. The result may not be the most efficient plan mathematically, but it comes close, and drivers will run it. (For more on how different route shapes are planned, see route types and how they’re optimized differently.)

Planning by hand still holds up on a small scale. Richard said a person planning 10 drops will usually land within 10 or 15% of an optimal algorithm. Give that person a thousand items across ten vans, and a poor call can mean a 100 to 200% difference.

Capturing local driver knowledge in route planning

One Adiona customer in Western Australia won't send drivers down a certain highway at night because there are too many sheep, and drivers have hit them. "There's no data source that I've been able to find that has a sheep counter on it," Richard said.

Adiona's route suggestions are advisory, and the people running the fleet always have the final say. When the same changes keep coming up, Adiona's models pick up the pattern and build it into future plans.

Experienced drivers are retiring and taking that knowledge with them. Of a driver with 20 or 30 years on the road, Richard said: "Good luck replacing that person without some automation and … without AI."

What a ride-along reveals about delivery driver burnout

On one of his first ridealongs, Richard went out with a contractor who routinely worked 11 to 13 hour days, five days a week. He hadn't taken a holiday in three years. His route was so complicated, with Post-it notes of customer phone numbers and door codes all over the truck, that nobody else could cover it.

That day shaped one of Adiona's principles. If a project won't cut emissions and won't make people's lives better, Adiona doesn't take it on.

Cutting fleet costs first, emissions second

"Most companies, they don't have money specifically for sustainability," Richard said. A few, like IKEA and PepsiCo, invest heavily in cutting emissions, but they're the exception. So Adiona leads with cost: "It is very much financial first, emissions second."

The biggest lever is fleet size. Adiona's solution treats a new vehicle as the most expensive option, because "adding a new vehicle to the fleet to do work is going to be the highest single cost that you can encounter." Clients typically do the same work with 10 to 20% fewer vehicles. The money saved on a truck they no longer run can go toward replacing others with electric vans.

On one Coca-Cola fleet in Australia, Adiona cut time on the road by 48% and distance by 36%, with emissions falling in line with the distance.

Why Adiona only takes on sustainable logistics projects

In Adiona's early days, companies wanted Richard and his co-founder to optimize sales cycles and where products sit on supermarket shelves. They passed, because that work had no sustainability or social benefit, and moved into logistics.

FAQs: delivery drivers and route optimization

Why do drivers ignore optimized routes?
Classic algorithms produce overlapping "flower petal" routes that look wrong to the people driving them. Adiona blends optimization with driver preferences, traffic and weather, so drivers follow the plan.

How many fewer vehicles can route optimization save?
Adiona clients typically do the same work with 10 to 20% fewer vehicles.

Do you need GPS data to optimize routes?
Usually not. Richard said GPS data is often patchy and hard to clean, and proof-of-delivery timestamps tend to be more useful.

Who is Richard Savoie?
Richard is the co-founder and CEO of Adiona Tech, a Sydney company that builds AI route optimization software for supply chains and fleets.

Related on Adiona