on “pilot” or experimental services

An amazing story from Norfolk, Virginia:

NORFOLK, Va. — Hampton Roads Transit is dropping a trial bus service to the Norfolk International Airport due to low ridership.

The agency said Thursday that the pilot program will make its last runs on April 14. The last bus will leave the airport at 11 p.m.

The hourly service began Feb. 5 and originally was to run for a month. The agency says it extended the service a month but ridership remained weak.

Hampton Roads Transit president and CEO William Harrell says the airport service was a good idea but it needs some work.

I have no basis for an informed opinion about the Norfolk airport service, but I would never recommend trying any transit service for a month, or even two.  Any transit service needs time to find its market.   My usual advice is that you shouldn't try an experimental service unless you're prepared to run it for a year, not only so that it can build ridership but also so that you can see how its performance varies with the seasons.  

When it comes to larger network redesigns, in which many route changes occur at once, the message for elected officials is even more challenging.  Redesigns are simply not reversible.  They have to run a full year before you can really assess them fairly, and by then, so many people are used to the new patterns of service that reverting to the old ones would be as disruptive as the intial restructuring was.  If an elected official needs to believe that a restructuring can be done "experimentally," I advise them to vote against it.  Obviously, a year after a restructuring, you'll tinker with things based on the ridership, and even sooner you may have to do small adjustments to handle running time or overloading problems. You'll make larger improvements after you've run the redesign for a year.  But there's no going back to where you were, and in a good redesign, after a year, few people want to.


15 Responses to on “pilot” or experimental services

  1. BBnet3000 April 5, 2012 at 7:58 pm #

    Most people would not have even heard about a new service to the airport in a month, much less heard about it and then had a reason to fly.

  2. Wad April 5, 2012 at 10:51 pm #

    I can conjecture why the service failed, and why keeping the line over a year would not have made a difference for the route.
    It’s an hourly service.
    Jarrett has said Frequency Is Freedom.
    Well, this would be the opposite.
    There is no excuse whatsoever for any transit agency to offer hourly service. I would even put the floor at 30 minutes.
    A modern society cannot function on a service that shows up once an hour. An hourly service might have been impressive when the legacy transportation technology was the mule, but most modern societies have long grown out of such primitive conditions.
    Since transit riders have something to do at the end of their trip, it is highly unlikely that a single hourly trip can align a person’s start time, travel time and arrival time in a useful manner.
    An airport bus service has two potential users: Airline passengers and airport employees. Both have very time-sensitive schedules to arrive at the airport, and the low frequency limited its potential rider base.
    This doesn’t just apply to airport services. This applies to all services. If you ever ride an hourly bus, you’ll see the vehicle mostly empty but if there are any riders, it’s very likely they’ll be elderly.

  3. Jonathan April 6, 2012 at 1:40 am #

    I guess the message of ‘reversibility’ that a politician wants to hear on behalf of the people is that: ‘If you, the transit advisor speaking to me, are an incompetent idiot and have hashed this redesign in cataclysmic fashion, we can still go back to what we were doing before’. And I guess the message that should be taken away from that is that if you are a politician who comes to believe that based on the outcomes do not wait the year.

  4. myna lee johnstone April 6, 2012 at 5:03 am #

    it is i suspect, just auto culture blaring its loud horn
    the Canada Line to YVR airport is superb yet thousands of people drive up, get dropped off or park.
    Ignorance and car culture prevail.
    Even when it costs more and is really inconvenient,especially so to those between planes stepping our for a breath of fresh? air.

  5. Jonathan Hammond April 6, 2012 at 8:16 am #

    Oh God! I grew up in Norfolk and that is even more idiotic than I’m accustomed to with Hampton Roads Transit.
    …Sorry for the strong language. Familiarity breeds contempt. This said, I presume “Wad” is probably right: in my experience, they have very extensive and very infrequent coverage.

  6. EN57 April 6, 2012 at 8:21 am #

    I’m not making any comments on the specific Norfolk situation. I have elsewhere observed that transit agencies and politicians tend to “trial” things they really don’t want to provide – but are only doing so to indulge some particular business or community interests. A real transit initiative has a solid business case behind it – taking into account demand, a reasonable patronage ramp up over time – as well as ensuring the service is designed to meet the needs of its intended travel market and will integrate with the wider network. One to two months’ trial doesn’t make sense for a new transit product, for the reasons Jarrett outlines.
    Wad, even hourly and two-hourly bus services attract patronage where there is demand, and bus services can and do graduate from hourly to higher levels of service on the basis of passenger boardings. If the service was for airport employees, it might have been timetabled around the main shift start times. For air passengers, services might have been timetabled more frequently around the busiest plane arrival and departure times, rather than just hourly over the whole day. You might start off with an initial targeted timetable to establish a customer base, before increasing to a 30min all-day service later on.
    Where the business case supports a frequent service on an otherwise untested market, starting off with 5 or 10 minute headways gives the opportunity (over a year) to identify the actual passenger demand profile under ideal service frequency conditions. Resources can later be saved by cutting out the trips or operating hours where there is no actual demonstrated need. In this case, while the agency might be internally trialing the initial frequency of the service, it has otherwise committed to its on-going provision. At least the agency can’t be accused of not giving the service a go.
    Myna, the train to YVR is still new!! Wait another few years for the population and traffic congestion to increase a little more – the train line is an investment for the longer term.

  7. EngineerScotty April 6, 2012 at 10:41 am #

    In many cases, the primary users of transit to the airport are the employees–unless the airport’s transit connection is really good. For many travellers, journeys to and fro the airport are infrequent trips, often made with baggage. And there are plenty of other options besides scheduled public transit and driving; most airports have a plethora of taxis, airport shuttles, and other dedicated services–quite a few of which are transit-like in that more than one passenger or group of passengers is in the vehicle–optimized for travelers.
    (One urban legend I’ve heard, which may be true but I don’t know so I’ll ask rather than tell, is that since airports are under FAA jurisdiction, private transportation services are able to serve it without paying heed to local regulations that might otherwise encumber such services. There are plenty of airport shuttles and the like serving Portland International which have multiple stops besides the airport, but which only sell tickets to and from the airport. You can’t use an airport shuttle to get from Albany to Salem, for instance, even though many of the same shuttle services stop in both on the way to PDX…)

  8. david vartanoff April 6, 2012 at 2:11 pm #

    In my experience I as transit geek may know some new service exists; most regular civilians don’t. It takes many months for most folks to have heard from friends/co-workers that a new service exists and is useful. A transit agency which deploys an experiment for less than a year is IMHO NOT interested in making it work.

  9. anonymouse April 8, 2012 at 1:13 pm #

    I think EN57 captured it perfectly: “At least the agency can’t be accused of not giving the service a go.” It sounds like a CYA move to me more than anything else. This sort of thing is what happens when, for example, local politicians demand a service that the transit agency already knows will be a failure.

  10. Wad April 9, 2012 at 2:14 am #

    @EN57, hourly service doesn’t cut it for anyone who has to be somewhere on a schedule.
    The Tidewater metropolitan area, which HRT serves, is home to 1.5 million people. This is inadequate for such a large population and it’s hard to infer meaningful data from low frequencies. Then again, HRT somehow built itself a light rail system, so it shouldn’t skimp on bus service.
    Another challenge I see for Route 7, the affected service in question.
    This isn’t a premium CBD-to-airport service. If there were few stops, then the lack of access points could explain the failure. In this case, it’s a regular local bus line between downtown Norfolk and the airport.
    This poses its own set of problems. It might be a challenge to connect to this service from the rest of the system, and the routing is likely the quickest for the transit agency to get to the airport, not necessarily one that jogs to catch the airport’s ridership, which would likely be employees.

  11. Clark in Vancouver April 9, 2012 at 11:48 pm #

    So true. It takes time for people to hear about new things. Many people, especially in a big city, just don’t bother following the news.
    I cycle downtown almost everyday but when they put in a separated bike lane on the Dunsmuir viaduct, I didn’t find out about it for five months. I had seen some sort of concrete thing at the start of the viaduct but didn’t know what it was and just continued along my usual (slower and more hilly) way.
    The Skytrain to the airport is amazing. I had a friend visit and he was so used to not bothering taking transit to an airport, because of bad service where he’s from, that he was going to take a taxi there. I suggested the Canada Line but he thought a taxi would be simpler. I had to show him on paper that it would be a single bus for five blocks and then a single subway to the airport. One ticket. He couldn’t believe that it could be so easy and cheap. It takes awhile to get used to quality transit.

  12. cph April 11, 2012 at 10:14 am #

    I looked at the #7 route on http://www.gohrt.com/routes/route-7.pdf
    I was expecting some sort of express service to/from the airport, but the #7 is just a local bus, in fact sharing much of its route with existing #3 and #8.
    Trip time from airport to Downtown Norfolk is about 45 minutes, slower than a cab or shuttle….
    I haven’t followed the backstory (not being from the area), but my guess is that this service was implemented due to political pressure (“Why don’t we have a bus to the airport?!”) In fact, many similarly sized bus systems lack service to airport terminals, even though the airport is relatively close to the city.
    What is the market? People in Norfolk taking a flight? Most would probably drive, get dropped off, use a taxi or other service before considering an hourly local bus.
    People flying into Norfolk? Again, most of them would also choose to rent a car, use taxis or have someone (Family member, friend, hotel shuttle, etc.) pick them up.
    Airport/airline employees? Maybe, if the route fit their work schedule, and didn’t take too much longer than driving. Perhaps this is the market that local transit should concentrate on. If so, routes need to be developed that serve the greatest number of employees.
    In any event, I agree that any “pilot” or “experimental” service requires at least a year or two so that travel patterns can be established.

  13. Nathanael April 18, 2012 at 7:47 am #

    What really helps with network redesigns is spare money.
    If you have spare money, you can build your new design and leave the old system running simultaneously. Like LA did when it added the big grid of “Metro Rapid” to the existing “Metro Local” buses.
    After a few years, with majorities of people riding the Rapids, you can start talking about dropping now-unpopular local buses.
    The problem most transit agencies have is lack of spare money. If you have spare money you can pit new vs. old and let the people vote with their feet. But if you don’t, you can’t afford to.

  14. Carl April 23, 2012 at 9:05 pm #

    I tend only to favor “trial” or “pilot” services if the chance of success is remote, but politcal reasons force “trying” the service anyway. Airport service is a classic one that Board members or the general public vastly overestimate the potential demand, so perhaps that happened here (Norfolk International isn’t exactly a major hub.)
    So what is the purpose? Try the route out because you legitimately expect it to work or because you have to demonstrate to others what you already know? If the latter, keep it short to keep it cheap.
    One note for fairness: effectiveness thresholds for pilot programs should be lower than permanent service.

  15. Carl April 23, 2012 at 9:22 pm #

    Not a big fan of full redesigns. A redesign on paper is a good tool and keeps you from purely reactive route planning, but the implementation should be incremental. For one, you’ll learn new things in the increments that will adjust the end goal. And redesigns of any frequency at all are a reliability failure for the customer. Transit needs to be something a person can count on.