In our backyard is the
Innisfil Transit experiment (Ubers as transit) -- kind of a rebirth of DIAL-A-BUS in a Uber format.
I think that experiment is being done quite inefficiently. If they are hell bent on doing this, then they should incentivize larger pooled vehicles as part of this experiment -- even the city purchasing them for city-trained Uber drivers.
Say, 7 to 10 person minivans (sub-$100K stuff). Innisfil is using some 6-person vehicles already but there's just not enough of them. So incentivize the bigvan:smallcar ratio! Even city-paid vans for some select city-trained Uber drivers -- would potentially pay for themselves when used on peak pool routes in reduced subsidies to Uber.
Then add one bus route along the main spine once enough ridership on pooled Ubers are occuring along a route, to take some load off.
They are
spending a bit more than expected on the Innisfil Transit experiment than if they did the original proposed 3 bus routes. While much more convenient than those original 3 bus routes for the users, and it has given some boom to Innisfil economy (
some local people apparently happy earning money as Uber drivers -- there is a $100 bonus incentivized here) -- there are unintended consequences such as more cars on roads, and the attendant damage to environment.
Electrification + Inclusion of bigger vehicles is probably key in future trials of these kinds of experiments (by any rideshare system, not just Uber, even one that a municipal government may roll-their-own in the future). The solo vehicles will still be needed offpeak and rural feeder but the larger Uberpool vehicles would ply the high ridership routes. And other optimizations such as requiring people to walk 1-2 blocks to the corner of the nearest feeder road unless they've applied for an accessibility/senior exception. Uber automatically tells people to walk to a more efficient stop or pre-defined stops in some cities (e.g. Las Vegas), so the software infrastructure to handle this efficiency increases is already invented.
Many optimizations can be be utilized in a good efficient transit-rideshare system that might be 3x-5x more efficient passenger-mile than Innisfil Transit, but a good focus must be had on deploying attractive transit (e.g. frequent bus in bus lane, BRT, LRT, metro, etc) once ridership becomes 0dense enough along a specific corridor.
Rideshare doesn't scale well beyond rural-ish, first miles, last miles, etc. There is a clear moral hazard here -- becoming addicted to door-to-door transportation. Pollution. Political pressures. Resistance to Uber's will (e.g. enforcement of certain rules that increase efficiency). Mentality against combining transit+ridershare as being one non-mutually-exclusive problem. Etc.
But, as these experiments will continue at least elsewhere in the world, in a probable relentless way, it is humankind responsibility to push for optimizations. It is likely still an option in solving transportation problems if the devil-in-details are adjusted properly.
So it is futile to dismiss these experiments as ridiculous, even as I am a transit advocate worried about potential environmental impact. Rather, I have to think pragmatically and realistically. How do we integrate municipal ridehail+transit in an efficient way?
In 10, 20 and 50 years, future descendants of Innisfil Transit experiment will probably include:
(A) municipal large-vehicle transit routes along the main spines;
(B) larger pooled ridehail along the popular ridehail journey routes;
(C) driverless operation for the smaller vehicles to cut costs;
(D) electrification requirements to cut carbon;
(E) incentivization/penalization to prevents empty vehicles going over long distances (deadheading);
(F) Incentivization of prepositioning driverless vehicles in unused resident driveways (whether owned by them, by other, or by city) nearer ridehail clusters, to reduce congestion and deadheading;
(G) Incentivization of a reasonable dwell period (~60 seconds) at peak at major stops to fill vehicles.full of riders who ridehail moments later;
(H) other efficiencies now made possible by mathematical magic on historical ridehail data.
Theoretically, GO Transit / Metrolinx could someday attempt a rebirth a DIAL-A-BUS equivalent in this modern ridehail era -- incorporating such efficiencies. If that ever happens, Metrolinx should not spend a Presto-style amounts of money to try to reinvent Uber from scratch, but partnering with a more domestic ridehail system (even one that comes out later thanks to driverless vehicles). To serve low-density and more rural areas outside urban cores (Toronto/Mississauga/Scarborough/etc.).
More realistically it would probably be other firms that does this sort of stuff, but relevant here as a "DIAL-A-BUS reborn" thought exercise.
www.channel4.com
www.kitchenerpost.ca
