I knew about Uber and their impressive software optimising the booking of Uber car and then smoothing the path by providing driving directions based on traffic conditions picked up by Uber cars in the area. I did not expected that software was also behind Uber services in taking pizzas to my home.
As reported in a Wired article, Uber is trying to get accurate forecast on the time it takes a restaurant to prepare an order and then the time it will take to deliver it to my home. By accumulating data, as an example, Uber has detected that restaurants in a certain area in San Francisco are some 20% faster in preparing Pad Thai (for the record, that is in the Mission District area with an average time from order to delivery at the biker that will take it to your home of 12 minutes and 36 seconds).
Now it might seem a bit crazy to try getting such an accurate estimate but that is what can make you, as a customer, choose Uber over… Grubhub, DoorDash and so on. Knowing when an order will be ready let Uber schedule the right rider to be there exactly when the order will be ready for collection. That is both good for the rider that will not have to waste time waiting and for the client that will get its pizza steaming hot. It brings efficiency into the value chain.
The customer uses the Uber app to select the menu items and gets an estimate, to the minute on when the order will be knocking at her door (and, like for cars, she has the opportunity of monitoring the pizza getting closer to her home).
To achieve this kind of accuracy (on the average, I cannot believe they can be accurate to the minute on every single order) Uber has to take into account a variety of parameters, including modelling the city landscape and the meteo conditions (since, in turns, they affect the traffic).
In areas with high order density they are able to combine order and delivery, making it for a better revenue to the rider and passing some of the savings on to the customers.
Clearly Uber is not alone. As soon as someone gets an idea that proves effectives, others will follow suit. DoorDash is also using math, artificial intelligence, and mathematicians (a bit of natural intelligence is still needed), to increase their business efficiency. They found out that orders flow based on sunset time, not on absolute time. You get peak orders volume later in Summer than in Winter, and on Friday’s Pad Thai preparation takes two extra minutes (because of congestion).
So far, in the US, Grubhub keeps leading in terms of market share, but their shares are rapidly dropping to the advantage of DoorDash with UberEats coming in 3rd.
The Digital Transformation is driving the race, and in a way it is surprising since we are talking about pizzas and bikes. Yet, bits are finding their way and reshaping the business. Had someone told me 50 years ago when I started University that the math I would learn would become crucial in the pizza business I would have felt they were pulling my legs….
One comment
Pingback: AI is becoming a marketing word and it’s a pity -IV – IEEE Future Directions