This involves understanding how our deployed services have had a positive impact on local bus routes; learning how citizens’ mobility patterns have adapted and become more fluid in line with these changes.
With this in mind, here are a few quick facts about our operations to date:
This was the first service that we integrated within a public-transit network. We substituted a classic line, which only ran 6 trips per day, for an on-demand system with longer working hours. Within a couple of months, the system scaled from an average of 5 passengers a day to 14 passengers. Almost two years on, daily patronage is reaching 30 passengers a day, this means ridership has multiplied by 6 since we started.
The line connects an isolated neighborhood with a main suburban railway line. Through passenger surveys we have learned that three main rider profiles exist: teenagers living in the neighbourhood and going to highschool, domestic workers without a car who are employed by households within the neighborhood, and couples who have to share one vehicle between two people.
When these users are asked how they managed these trips before, they tell us that they relied on somebody else to pick them up.
Since last November 2018, we have collaborated with the public transport authority in Oulu to offer a service that connects the city with the northern town of Haukipudas (20.000 inhabitants). The service covers an area of 8 km2 by using 14-seater minibuses. After a few months of operation, we found out that users wait an average of 11 minutes when they request a ride and spend 14 minutes onboard.
We have rapidly reached figures between 50 and 80 passengers a day, in an area that previously had no service at all. When mapping demand in relation to time, origin, and destination, we observed that every day there is a peak in users, who all get on at the stop next to the school, at around 3pm.
In Vallirana, a suburban town of 15.000 inhabitants, one minibus was previously used along 4 different routes. This meant that at any stop, bus frequency could be up to 1 hour and 15 minutes, sometimes even more.
With the deployment of Shotl, we converted these 4 lines into a cloud of 51 stops. We then added 35 extra stops to increase coverage. By building routes in real-time and in accordance to user demand, the average waiting time went down to just 20 minutes. This had the positive effect of increasing daily ridership from 20 to 60.
We realized that on this route, especially in the morning, the percentage of users requesting rides through the call center function - and not through the app - was higher than usual. After checking in with service users on the morning route, we observed that a high percentage of seniors were using the line. “I used to go to the town center only once a week, planning my shopping and visits with friends accordingly. Now, I can come back to the center with ease!”
All three of these cases have one common factor: Shotl deployments are making life easier for commuters who don’t have a private car, this gives them greater access to education, commerce, leisure, and employment.
Pollution, congestion and changing mobility are challenging the car’s dominion over cities. Amsterdam is the latest taking bold steps to reclaim the streets for people, with potentially huge benefits.
Machine Learning (ML) is a field of Artificial Intelligence (AI) which focuses on the development of techniques that allow computers to learn how to perform a specific task.