On-Demand Service Design for a Heavenly Mobility Future
June 24, 2019 | Company Blog
The title of a report by the Greenline Institute puts the question surrounding autonomous mobility services starkly: “Autonomous Vehicle Heaven or Hell? Creating a Transportation Revolution that Benefits All.”
The authors point out all the potential benefits of shared electric autonomous vehicles: cleaner skies, clearer streets, and affordable transport that can take people door to door more efficiently and at a lower cost than that of a private auto.
The key word is “potential,” as the report says, “If left to the free market without adequate regulation, we can expect a ‘hell’ scenario dominated by personally-owned autonomous vehicles that are only available to those who can afford them, while further congesting our streets and polluting our air, leaving others to cope with worse traffic, longer commutes, and under-resourced public transit.”
Their findings echo research from Zurich’s technical university, ETH, that recommend the prohibition of private ownership of private vehicles, and from the University of California at Santa Cruz that found private AVs would double traffic in cities.
But for shared electric autonomous vehicles be adopted at a scale that will reduce traffic and pollution, they must be able to operate more efficiently than alternatives, including private human-driven autos. This is no given. In the early days of ridehailing, companies like Uber and Lyft promised that their services would reduce congestion in cities. Instead, the opposite has happened. Ridehailing services have added 5.7 billion miles of driving in U.S. cities alone, and traffic is 180 percent worse.
While regulation may keep private AVs off the streets, it won’t make shared fleets more efficient. Cities’ experience with ridehailing services has proven that new technology and services alone don’t solve congestion and pollution problems.
It has been proven that shared services can be more efficient and convenient than private autos when they are properly designed and work in concert with public transit. Studies by MIT and the University of Texas have found that shared services can handle as much as 98 percent of ride demand with 75–90 percent fewer vehicles, with a faster door-to-door experience than a solo commute.
The critical success factors in achieving these outcomes are the service design and fleet orchestration. The MIT study built theoretical algorithms that could instantly process demands from riders, the locations and destinations, all vehicle locations and capacities, all possible trips, and offer shared rides that result in minimal excess ride times due to sharing.
Bestmile has conducted similar research using its own Fleet Orchestration Platform and found similar results — a 10x improvement in fleet efficiency with algorithms that manage supply and demand in real time, and machine learning to continually improve performance.
Service design is a professional service Bestmile offers to help mobility providers plan and optimize new services prior to deployment to deliver heavenly mobility. Service design uses realistic demand and traffic data and creating virtual fleets to simulate and test multiple parameters such as fleet size, vehicle locations, vehicle capacities, fuel levels to find the right balance of supply and demand. Testing in a virtual environment can be far less costly and risky and can deliver much more accurate results than trial or pilot projects with small sample sizes and service areas.
To learn more, register for the webinar, On-Demand Mobility Service Design: Making Fleets 10x More Efficient.