For many years, i have envisioned a system of public transportation where the vehicles are owned by the community and made available to all members of the community. My original idea, ambitious as it is, was to have computers controlling a fleet of five-passenger vehicles and some sort of central dispatch system to coordinate travel plans for all the passengers, organizing the vehicles to minimize fuel usage by intelligently grouping sets of passengers with overlapping segments in their plans. I realize that is a lofty goal, and there are a lot of incremental achievements that will move our society toward such a system. With that in mind, i recently had a conversation with a friend about this.
My friend has a unique perspective on the company that is responsible for managing public transit in Tampa. I won’t get into detail about it here, but suffice it to say there is a great deal of waste involved in the current system. My friend and i discussed some of the ways we might be able to evolve into a better, less wasteful solution. Since we can’t just step into a brave new world where everything is ideal, we must break that step up into small improvements. We must also make concessions where radical change is not yet feasible.
The first concession we agreed on was the computer driver. While the DARPA challenge has inspired significant advances in robot vision, collision detection, and vehicle control, the technology still has a long way to go before passengers are willing to trust a computer with their lives. We’re just not there yet. Frankly, i believe the technology is mature enough. The key barrier is public perception. Despite the simple fact that computers control our bank accounts, fly our commercial aircraft (except takeoff and landing), and generally do a bang up job the vast majority of the time, people continue to be skeptical.
When we removed the computer driver requirement, the system simplified substantially. Vehicles smaller than a bus, say 5-8 passengers, can be highly efficient, especially if they can be filled to capacity. Take a look at any bus in Tampa outside of rush hour, and you’ll see a lot of empty seats. Instead of driving fixed routes on a fixed schedule, smaller vehicles individually tasked with a custom set of waypoints defined by the real-time needs of the passengers can solve the same problem for less money. In fact, the only real capital expense is the vehicles themselves. We already have a battalion of licensed drivers, some of whom sit on the bench on call. What we don’t have is the central dispatch system.
Now, if we were to build a web/mobile app to allow passengers to define a set of waypoints representing a travel plan, and then organize the plans based on real-time position and occupancy information from the vehicles, we would be able to organize the vehicles and communicate with the drivers to achieve the desired result. When a passenger creates a new travel plan, it would be matched against existing plans to determine which vehicle would be best suited to handle the request. The matching vehicle’s plan would immediately be updated to incorporate the new passenger, and the driver would be notified.