Pizza places have a dynamic load that is only somewhat predictable. They would greatly benefit from instantaneously dynamic availability of drivers. There's no reason why the Uber/DoorDash/etc driver can't pick up multiple pizza orders from your pizza shop and deliver them without providing service to another company for the duration of that delivery. The benefit is that now that driver doesn't have to come all the way 5 miles back to your pizza shop, because maybe McDonald's a block away has a delivery need. And someone else who just delivered a brake pad to the auto shop next to your pizza shop can pick up the next pizza deliveries.

I think the primary benefit of having your own driver is that you get to put your branding on the car and the driver's uniform. It's not clear to me how much seeing that little "Pizza Hut" car topper impacts sales.

Balancing across multiple restaurants is only useful if they have different peaks during the day. It’s mostly useless when they all share similar profiles as you need to employ more people for the peak load at lower efficiency.

Restaurants can also make use of drivers they directly employ durning downtime to do other stuff like sweep up.

We live in an world where UPS famously saved hundreds of millions of dollars in labor and fuel costs by favoring right hand turns in their routing. Tiny improvements at scale are a real savings both in cost and environmental impact. Not having to return to the same single central hub after each delivery alone would be a massive cost reduction.

https://www.ups.com/us/en/services/knowledge-center/article....

Once you start optimizing across large numbers of drivers you end up dispatching the same driver to the same restaurant to do multiple pickups anyway, which then need to be delivered quickly. Outside of peak times having fewer drivers than restaurants seems like a net win, but restaurants can make use of drivers when their not making deliveries. This also means their in the restaurant and thus lowers delivery times.

Believe me I understand why it seems possible, but this is one of those cases where real world data doesn’t fit abstract models very well. One example is during peak times both the restaurant and it’s drivers get overwhelmed so preparation time is increasing. Another example is if their returning to the same location they can easily return empty insulated boxes, but that’s problematic if their doing pickup from 10+ restaurants a shift.

Wouldn't that imply an algorithm weakness in this case? It seems like the wider the network and the more driver's you have the more optimal routes can be given algorithms good enough.

Pickup and delivery is not simply instantaneous events at the end of a route. For example, you need different processes for in house drivers vs 3rd party drivers. Many pizza places benefit from gathering orders into insulated containers for delivery which drivers then pickup an return.

On the other hand someone showing up at potentially hundreds of restaurants is hardly going to know each of their systems very well or be recognized on sight.

I was thinking about that the other day, and it's actually a huge difference in the delivery model when you move from "single pickup point, multiple delivery points" to "multiple pickup and delivery points." The first is pizza, the second is a taxi, even if the taxi "passenger" is your order from Cheesecake Factory.

That same Pizza Hut driver can deliver multiple Pizza Hut orders per trip.