Commuter Solutions

TaaS-A-mazon: Transportation as a Service

Do we really understand what needs to happen to urban motorized transportation as it evolves further into Transportation as a Service (TaaS)? To deliver the Goldilocks world of tailored vehicles arriving within two or three minutes of demand we will need aggregation and matching services for large numbers of disparate, multi-owner fleets, far beyond what anyone, including Uber, provides today.

Many people see the household robotic vehicle replacing the current household vehicle. Some figure a reduction of 1.2 vehicles to one. Others project each household vehicle would be replaced by 1.2 robotic vehicles. Fundamentally these ratios represent a like-for-like trade, just without the steering wheel.

Other people think about the robotic taxi replacing the taxi and a robotic Uber replacing a human Uber driver. They willingly admit this might begin to nibble at transit – say 10 or 20 percent – but is still essentially a like-for-like trade, but with fewer drivers.

Of course Travis Kalanick sees a future driverless Uber replacing nearly all urban, automotive transportation. But even people hearing that do not really get it. Or believe it. It is like hearing the seas might rise by a meter and not really appreciating what that means.

Problem is, most people see the current modes – car, taxi, carshare, shuttle, bus – as just how transportation is meant to be divided up, as determined by a kind of DNA, or by the same imaginary book of rules that say bike-paths are anti-car. Some folks will admit some 15 percent shifts, but they do not see a total obliteration of taxi-car-bus-shuttle-carshare boundaries. They do not see that these modes just become different vehicle sizes or inter-mixable aggregations on appropriate routes from what would act as one massive common service.

Has iTunes and Amazon taught us nothing? People go now to Amazon and search for the product they want. People find that in addition to an Amazon warehouse full of products, there are thousands of other sellers hawking stuff there as well. Amazon is the marketing and pricing front end, the social hub for the buyers’ emails and “likes”, and the bank that escrows the money until the product is shipped.

Amazon sells trust and payment reliability more than it sells books or can openers. Many of us buy books there – including used ones – because we trust Amazon’s reputation management system. There are lots of resellers there, and most of us cannot remember if we ever used the same one twice. It’s all just Amazon.

How will people order a personal car trip in the future?

This suggests an analogous mobility aggregator. How might you order a personal car trip in 2035? First and foremost, you would have a profile reflecting normal daily use, for example:

Maximum fee: 40 cents per mile; Minimum fee: 25 cents per mile (to make sure not to get a beater); Preferred size: 2 seats (who wants to hold their groceries on their lap?); Ride sharing: no; Premium to meet schedule: yes; Maximum walk in meters: 200; Child seat: no; Cargo: yes; Cargo size: medium and Payment card: ****7221

Of course you could override these, but usually you would enter a destination time and place, then click to order and walk out the door for its arrival.

Importantly, a tailored vehicle that fits your immediate demand might be owned by a city agency, or perhaps by Uber, or by Cisco. As often as not it might belong to one of the several university, shopping or association affinity co-ops you happen to belong to.

When you order a trip to the airport, perhaps United Airlines will own the vehicle, since they co-manage a large autonomous shuttle fleet based at one of the emptied-out airport parking garages. Short trips might be assigned to one of several local operators who manage maintenance and cleaning of 30 to 100 robo-cars as a family business. Vehicles assigned for longer, intercity trips might be managed by the successor to Greyhound or Megabus, since those vehicles have different long-haul trip and maintenance properties.

But when you look at your smartphone, you are thinking only about the trip itself. You would care little about the brand of the vehicle or its provider and more about the provider’s current reputation on the app. Just as thinking only about the book you want from Amazon, you care little about the company that printed, sourced, picked, wrapped or shipped the book. Why would you care in 2035 whether the vehicle is forwarded from the municipality, Lyft or Wal-Mart, as long as you expect to be satisfied with the trip.

Could it be that Amazon has not announced an autonomous vehicle because they intend to become TaaS-A-mazon – a clearing house for everything with a brain on wheels?

One concern this disruptive threat implies is for progressive communities to find a way to protect transportation equity. We might not all be able to buy from Amazon, but we will all need to get somewhere. If transit agencies just wait this out, their services will be disrupted, ridership will drop and subsidies will become ever steeper for the remaining users – and those users will be the ones that could not make the switch. Commercial TaaS fleet operators would have no motivation to subsidize those riders, and neither would a TaaS-A-mazon.

Do you rely on online goods and services aggregators to initiate purchases for things such as books, hotels and vacations? Do you trust them? Would you prefer this for TaaS rather than directly joining and ordering from Zipcar, car2go, Lyft, etc.?

Please note that this article expresses the opinions of the author and does not reflect the views of Move Forward.



  • stephenspower
    5. February 2016 at 18:58

    This is great. The only thing I’d speculate to add to this is a box after ride sharing for “personal” or “public.” That is, “personal” would let you, say, carpool with three co-workers, each picked up in turn, while “public” would let you carpool with strangers headed in the same direction. I imagine the former could also offer a bonus for sharing the cost instead of each person pay for separate rides. The latter could do the same with, perhaps an override vote so the three in the vehicle aren’t made late by someone trying to jump on, but forcing the vehicle to go out of the way. The sharing option would also offer a discount to encourage it so other vehicles are freed up for other rides.

    Seeing as this will be used in the suburbs mostly for shuttling kids to and from games, events, schools and playdates, I wonder what security protocols could be added. It’d be also interesting to see if once drone cars are used for actual dates, what parental overrides could be added, such as “my daughter is not allowed to ride with this roster of boys.” Will done cars have streamable cameras so parents can make sure their kids aren’t doing anything illlicit in a drone car so it doesn’t become a driverless party bus.

  • Grush Niles
    8. February 2016 at 15:29

    Thanks for your comment.
    re sharing:
    I think the idea of personal vs public ridesharing preferences is interesting (personal being from among one’s network of ‘friends’ in the Facebook or Linkedin sense). Using this to share rides to work among a dense employment network is interesting, but in general, this would severely limit ridesharing or timeliness of trip starts.

    Don’t forget there could be an opportunity to “rate-your-ride-sharer”. In cases of two strangers sharing a ride (a more common occurrence than three or more) it would be possible to rate your willingness to share in future (without each other knowing). You could choose among, say: “absolutely, share again”; “it’s ok if it saves me money”; or “never again, thanks”. This would create an anonymous network of which people who are willing to share with which others. Such people might socialize or not, but what would happen is that as the network enriches, ridesharing would go up and discomfort would drop.

    re security:
    I suspect there will be a lot of security protocols and I assume cameras will be used for security reasons (as cameras are in taxis now). But I think teen behavior needs parental coaching, trust, openness, and understanding more than cameras. If a teen is prevented from a behaviour because of a camera in a car, then that behavior will happen outside the car.
    Bern Grush

  • stephenspower
    8. February 2016 at 18:16

    That’s a great idea, re rating your ride sharer, as well as the notion of forming whole new social networks through carpooling. Of course, I give the system about a week before carpooling becomes a euphamism for “adulterous hooking up.” At which point divorce attorneys will look into whether their footage can be subpoenaed by the cuckolded.

    You’re absolutely right about teen behavior. Teens might also be hesitant to use these cars if the movements can be mapped to one location.

  • Grush Niles
    8. February 2016 at 22:35

    I have to stifle a head-shaking laugh, here Stephen. I have been in Transportation demand management for about 15 years and the leading concern that keeps “rising” in related discussions is men’s predilections to sexual chicanery (never women’s, which by definition must occur with similar frequency!). In 2004, I was asked while speaking at a parking conference whether the GPS in my automatic parking meter would tell his wife if he were having an affair. In 2006, I was asked on the radio (by a highly respected male host) whether, if he used my GPS road tolling system, his wife could see when he visited a strip bar. These questions continued since. Now your concern that carpooling will lead to adultery and a new spate of work for divorce lawyers is new, even for me, but the drift is unchanged. Sadly, we men let our “membership” colour almost everything. One thing is clear: when many men contemplate the value of privacy they are contemplating the freedom to cheat on their spouses. Even more ironic is the fact that the remaining few laws against adultery in North America were written by men in the first place. Perhaps it is time we made up our minds.

    On the other hand, I have heard that autonomous vehicles will be taking all the driving jobs away. If you know someone just starting out and trying to decide between taxi driver and divorce lawyer, encourage them to apply to law school. Bern

  • opbrid
    10. February 2016 at 9:10

    Hard to say. Might be similar to Skyscanner where I first look for flights, which compares and redirects to the final site. Though taas-a-mazon would be more convenient. Could be that I prefer a certain company in my neck of the woods, because they keep the cars cleaner or better maintained. In any case, this is a detail, and by then, the user interface may well be a chip embedded in my brain.

    BTW Stephenspower, I really like the idea of a “driverless party bus”!

  • Paul Minett
    11. February 2016 at 4:17

    Interesting article. Yes, I do rely on online goods and services aggregators – specifically books, hotels, airline flights. For hotels and airline flights, the reputation of the actual provider is also in my mind. For books I don’t really care – but I am usually there looking for a book that has been recommended to me by someone else, I do not use them for ideas of which books to buy. My point is that ‘trust’ is relative. The magnitude of disappointment potential is much lower for a book order than for a flight.

    I find it interesting that you condemn us all for our mental models and then perpetuate one of your own. Why in 2035 would I even think in terms of ordering a ‘personal car trip’? If I had a profile such as you suggest, why would the maximum cost per km be the most important criteria? I am not saying you are wrong, just that you seem to perpetuate the current model and way of thinking.

    I used to (and still do sometimes) get entertainment content from an aggregator – my TV aggregated products from local TV providers who themselves were aggregators – and now all of a sudden this model is being disrupted by a completely surprising (for me) offer of content streamed to my phone. Both my TV and the local station are becoming obsolete.

    Due to the silly limit of 2000 characters, this comment is continued in the next comment, above.

  • Paul Minett
    11. February 2016 at 4:18

    Continued from the previous comment, below:

    Cities in the future will work much better when there is a lower overall incidence of concurrent vehicle trips that exceed, during peak, the capacity of the road system. In some way, we hope that communities can find the means to resolve excess peak demand. My mental model for 2035 says that when I desire to be physically in a different location at a given time (and an expectation for my return as well), and I will communicate that need to some sort of service – perhaps quite similar to the water or electricity utilities that deliver, well, water and electricity. If I want to travel alone I might expect to pay a higher price for the journey. It is hard to say if I will have several providers available to me, as I do with my electricity – or just one, as I do for my water. I have trouble figuring out if I care if it is an expanded Lyft offering that I get, or a higher level service that aggregates Lyft with other services – what will be important is that I trust the service to get me there and back.

  • Grush Niles
    14. February 2016 at 13:59

    Paul, this is not *our* mental model, rather it is the robocab model proposed by so many with an aggregator superimposed. We assumed that all the cars on the world would not be owned by one company (such as Uber) and that we users would tire of having to figure out which company could get a car to use fastest. Hence the aggregator. (Actually, the aggregator would arise just becasue someone could earn a living from it.)

    We do not think that max cost per km would be the most important criteria, it was simply one on the sample list. Its position does not indicate weight. For many now, cost per km is hardly the top motivator for our car choice.

    The mental model you ascribed to yourself is exactly the one we had intended to describe. In our profile example, the field *Ride sharing* was marked *no*, but it could have been marked *yes*. Even the field names in that sample are just examples. We were only after the aggregation concept and not intending to propose either SOV or ridesharing.