Transit and roads: the future won't be like the past

Baltimore Transit advocates ponder if a city transit agency would provide better service than a State owned one, while the MTA battles with equipment to make real time bus arrival information finally reliable in Baltimore; BC-DOT meanwhile proposes traffic actuated signals, additional turn lanes and parking restrictions for commuter cars on Boston Street while still punting the decision of making St Paul and Calvert Streets two way. Thus is the current reality of Baltimore transportation.

The cartoon says Buffalo, but it could say Baltimore or "transportation"

Those activities can strike an outside observer as rearranging the deck chairs on the Titanic, or the orchestra arguing about the next piece of music while the ship is sinking. Or as the famous projection of the depth of horse manure on the streets of Manhattan at a time when the horse was about to be replaced by automobiles. This type of fundamental failure to adjust to new challenges is proverbial and it is extra accentuated when the presence is already lagging behind what other cities do.

Missing the right moment to respond is easy, especially when whatever adjustment one would take has long lag. As in climate change. Or, in the case of transportation, the autonomous vehicle. Who should act first, the manufacturers, the insurances, the Feds, the states, the cities or private developers?

The low tech problem of "bunched buses" (#54 Local Link)

The issue lingered for years without much fanfare and only recently reached a fever pitch in the media. Still,  neither the MTA nor BC-DOT have developed plans for a future in which any of these entities would have created a new set of circumstances making current assumptions utterly obsolete. A State task force dealing with the legalities of the AV (MDOT's Connected and Automated Vehicles (CAV) Working Group) is the only sign that anybody pays attention. The group's August agenda doesn't convey any sense of urgency. This stands in stark contrast to other states which actively test AVs and have the proper laws already on their books. Virginia's parting Governor told techies and mobility execs in Arlington this spring that he wants to see Virginia up front when it comes to drones, AVs and other tech innovation:

“I want to own the land, the water and the sky,”  “We’re going to bury those other 49 states. Worthless!” Virginia Governor McAuliffe.

The AV won't only affect road planning but it will drastically change cities, land use planning and also transit. By now it is a pretty common assumption that the autonomous vehicle (AV) will upset everything transit planners currently take for granted. Danger lurks in places that don't come immediately to mind: For example the powerful transit union in which bus and train operators are organized may become obsolete. Pension plans for retiring operators may collapse when they the aren't any longer fueled by active operators and potentially drag agencies into a financial abyss, not unlike the one that America's automobile companies faced after the financial crisis.

Volvo AV used by Uber in Pittsburgh

Most transit agencies still plan their systems around  the ironclad facts that their biggest operating expense is the bus driver or that their guaranteed riders are those who have no car. Both of those assumption may prove wrong once the AV becomes widely available. Nelson Nygaard's Director of strategic planning, Jeff Tumlin predicts that in 20 years the bus driver will be just as obsolete as the elevator operator is today. Self operating subways have been in use for several years in Copenhagen. Just as the social compact with retired drivers collapses when there isn't a crop of new drivers propping up the pension funds, it will also collapse if private companies would begin offering driverless transit. If anyone with a van or bus can offer a cheap demand based transit option by eliminating the high cost of the driver, traditional transit agencies will lose their riders quickly. How much new technology can upset cost paradigms has been amply demonstrated by Uber and Lyft and the disintegration of the value of taxi medallions.

Sometimes it isn't even technology but just a new start-up which creates upheaval in transportation: Startup airlines managed to beat out legacy airlines simply by not having to deal with the high legacy cost and by some innovative deployment methods such as open seating which let Southwest get their planes into the air faster than the competition. For transit agencies to survive they need to think like Southwest Airlines. Be nice to their employees, care about customer service and innovate wherever possible. Sometimes those innovations are not based on technology, just common sense. In a transition period agencies need to offer some demand-based transit themselves, test AVs and collaborate with other share providers to link traditional transit service with "last mile" services. So far, the transit union ATU blocks all of this. Not so in Switzerland where a one year test has begun in real life conditions:

Self driving small bus in real city: Helsinki, Finland

The buses come from startup BestMile and will be operated by SwissPost transport subsidiary Car Postal. The company’s name hints at the purpose for these small, nine-person shuttles. They are designed to cover the “last mile,” the gap between a regular bus or metro stop and the passenger’s own front door.
Unlike Google’s private self-driving vehicles, BestMile’s focus is public transport. It also takes a different approach suited to a network of vehicles running on known routes: the network controls the buses “the same way a control tower does in an airport,” writes BestMile. The company is also working with the Swiss Federal Institute of Technology in Lausanne to improve the technology for better control and routing. (Fast Company)

A transit provider using self driving buses and vans will change so many transit paradigms at once that it is hard to imagine the full scope of ramifications. Certainly the neat divide between the personal car and transit would disappear once transit could offer demand-based service in addition to fixed-route, fixed-schedule transit.

In April, the UK's Transport Research Laboratory (TRL) completed its first test of an autonomous shuttle in Greenwich, a part of Greater London. A prototype driverless shuttle carried members of the public through a 2km route around the Greenwich Peninsula, using built-in sensors and autonomy technology to avoid obstacles. (CNet)

Gateway shuttle near London, England

Similarly, the folks at BC-DOT are caught in the old paradigms. They still take it for granted that more and more cars will push through the streets of Baltimore based on economic activity and growth assumptions in certain areas such as along Boston Street. In their most progressive mode, they prepare for epic fights about street space allocation for the various modes, bikes here, buses there, pedestrians somewhere else as part of "complete streets" policies. The matter is usually seen as a zero sum game where so much space can only be divvied up so many ways. But that is a false assumptions even before one thinks about AV's. Demand and supply are far from zero sum games even if space is static. More supply creates more demand, less supply reduces demand. Favoring one mode over an other will actually influence the mode split, in spite of the cherished illusion of transportation planners that they are mode "neutral" and only provide for a demand they first measured. In truth, transportation planners are far from being neutral, instead they reap exactly what they sowed. Their seed so far has automobile written all over its DNA.

Jarrett Walker's "geometry problem"

In the AV world BC-DOTs plans for Baltimore's streets would change even more drastically than MTA's for transit. For example, smart AVs can recognize signals without large physical overhead installations because the info about whether a signal is red or green would be transmitted as a WiFi code. Smart AVs recognize pedestrians and bicycles and the much loved protected bike lane would potentially no longer be needed. Most importantly, not every AV should be operated by an individual owner but instead be part of a fleet AV similar to bikeshare bikes or NextCar cars. Thus the AV wouldn't sit idle all day and be picked up after work on some parking lot like today's private car, but it would pick up the user on command. In motion the AV wouldn't act like a jerk and cut others off, cause unnecessary friction or crashes. Instead it would neatly fit into a platoon of vehicles moving along per some optimized algorithm in which many vehicles "speak" to each other and the surroundings. Experts are undecided, how much urban space would be freed up from parking and streets this way, how much capacity would be added on the given roadway lanes and how all of this would influence how users would use AVs. Once again, added supply will influence supply. Will people en masse send AVs  to get pizza? What impact would all this will have on mass transit? Transit guru Jarret Walker is afraid that the ride share AV will make traffic worse by killing transit and creating even more congestion.  Sure is, that all what is gospel today will be obsolete as soon as those AVs are out there in significant numbers and that there is an urgent need to shape this future.

Daimler Benz self driving test bus

When? Sooner than the prevailing slumber suggests. The future outcomes depend less on technology than on regulators and policy makers. Right now it isn't necessary or fruitful to accurately predict the future as if technology were an independent force.  For BC-DOT, transportation activists and the MTA it is urgent that they begin to design and plan a future which they, their employees, their customers and the people of Baltimore would find desirable, practical, equitable and convenient.

Klaus Philipsen, FAIA

Switzerland’s New Self-Driving Buses Will Probably Run Like Clockwork