Sunday, January 15, 2012

135> In Search of a Parking Space

The project in Amritsar has underlined some issues about PRT station design that would seem to merit a second look. The system under construction will have seven stations, and 200 vehicles. That comes to 28.6 vehicles per station.  The question, then, is, “Where do the vehicles go when not in use?” The illustration above should give pause. The structure and land use is huge, as are the number of bays, and even this is not enough to house a seventh of the vehicles. Note that there is no parking shown. How do the people get here? More importantly, where do cities get that kind of real estate and how much does it bump up system cost? Here are a few observations:

One of the classic features of PRT design is the off-line station, where empty vehicles wait to take passengers on their way. In the straight-line model favored by Dr. Edward Anderson (now embodied by ITNS, Skyweb Express, Skytran and others) the vehicles wait single-file, similar to a line of taxis. There is none of the inefficiency of backing up, but this advantage must be weighed against the fact that all departing vehicles must wait for the leading vehicle. The greater problem with this serial approach is that it is ill-suited for vehicle storage, as Amritsar exemplifies. For example, if a final occupied vehicle comes into a station that already has 28 empty vehicles in a line, the station would have to be 29 vehicles long to receive it. In the illustration below it can be seen that the tailing vehicle’s passengers cannot get out if there are no outbound passengers.

Two solutions come to mind for such systems. One is to simply have a storage facility. As usage declines vehicles would simply take themselves out of the system. The other is to separate the stations into two halves. There would be arriving and departing sections, separated by enough track to store the vehicles. Then, in the case of the last illustration, this could be the loading station only, with all waiting/stored vehicles already emptied. These are both simple enough solutions, but they involve significant system changes.

The saw-tooth parking approach embraced by ULTra, Mister and others, where the vehicles back out, can, in theory, be quite compact, at least in terms of fitting in normal rectangular building lots. Vehicles that are designed to be front loading can be arranged in a typical perpendicular parking configuration and spaced so close together as to be almost touching. This density would be limited only by turning radius and the dynamics of pedestrian movement in the boarding area.

The preferred embodiment may depend on what shape of space is available. For example, the very long, narrow space required for “straight-line” stations might frequently be readily available within existing easements. If it is not, however, and land needs to be procured, the straight line design might prove highly impractical. To further complicate matters there is the matter of visual impact. It might be that the stations or stored vehicles would be considered obtrusive.  

A couple of additional observations: Maintenance and/or recharging might be a factor to consider, as these activities might take vehicles out of service anyway, so some percentage of the fleet would have this additional place to be at night. Also, one interesting idea would be to simply lease office space and store vehicles in a office building or retail space. The space could even double as a private station for tenants. (They probably wouldn’t welcome the extra pedestrians of a public station) Another thought involves open automobile parking. Two of Amritsar’s stations are parking areas, and this may well prove typical. Could automobile parking and “pod” parking be the same? After all, it is after the auto traffic thins out that the PRT vehicles would be most needing spots. In the morning, when the cars roll in, the pods would be there waiting. Alternatively, PRT storage/station could be on the roof over covered parking, something the ULTra illustration would have probably included but for the promotional nature of the rendition.

My own view is that it is important to be as flexible as possible. That means, from a vehicle/track design standpoint, very tight turning radii both vertically and horizontally and very steep slope capability. This enables many configurations for both boarding and storage which are impossible otherwise, such as tightly packed three-dimensional storage arrays. A pod’s door placement is also an extremely important detail that I am still wrestling with. (In post 116 I designed several variations of the in-line type stations which utilize parallel, double-sided boarding, and require left and right side doors, although I have also played with front boarding designs)  

My suspicion is that, upon reviewing actual potential routes and station locations, city planners will generally conclude that these are the challenges that are the hardest and most expensive to conquer.  It does no good to have a system which, by the mile, sounds cheap, but then requires unexpected purchases of real estate, or requires track layouts that prove unacceptable to the communities involved. These are not details but rather the main challenges - The solutions that get weighed against (and can win against) other transit alternatives. The challenge is to have a practical answer for every case, rather than to force a city to search for alternative routes.  If a system won’t work in essentially every instance, it has little chance of ever becoming a pervasive network, and we all know that is where PRT will really shine…if it ever gets that chance.


Andrew F said...

Dan, isn't the obvious solution to just put in some sidings every so often to store/stage surplus vehicles? If the vehicles are 4 m long and can be parked nose to tail, it does not require much additional guideway to store 100 surplus vehicles throughout the network--you could even add charging functionality. The best place for such staging is just before the stations that expect the greatest surge loads--such as a bus or train station. Since the vehicles are all identical, random access is not necessary. I would keep these distinct from stations, to simplify their operation and to keep them no bigger than necessary.

I always thought this was a relatively straightforward problem.

Dan said...

Hi, Andrew… I will not say “Dan the Blogger Responds” this time because saying “Dan the Blogger” repeatedly doesn’t seem to make searching “Dan the Blogger” much more likely to refer to this blog!

Being a suburbanite, most of the traffic I encounter is on the highway or the adjacent feeders and cross-streets, where such storage would not be a problem at all. The really highly urbanized areas that I encounter that have really bad traffic tend to be in places where every square inch of space is already spoken for, either because the area is very old (Boston comes to mind) or has never been subject to any zoning or planning. (Houston) It’s not like every situation comes with a big boulevard to build over. Also, without public funding, ability to exercise eminent domain rights, or strong support from politicians these little problems will tend to get magnified.

To be honest, in most situations it really isn’t a big deal. It is like any other aspect of PRT. It probably isn’t a big deal if the speeds are 10% slower, or if there are is a bit more canopy, a bit more noise, or if stations need to be a bit bigger, or a bit farther apart, etc. None of it is a big deal. But these things add up. Having spent a good portion of my younger adult life planning and supervising construction projects, I know from experience that the “devil is in the details”, and that there will be problems enough that cannot be anticipated, let alone those that can. Also, it is the purpose of this web site to question these things. I am trying to push the boundaries – to explore the ways in which PRT can be an even more versatile transit solution. If it can be used in 85% of instances, why not 95%? There may be cases where several minor problems can be remedied by a single design solution. But it’s kind of hard to “kill two birds with a single stone” without going on the hunt in the first place.

Let me offer an example: On Unimodal’s old cover page, they showed the Skytran system with the main line on an upper level and the station’s feeder line below it. Most systems show the feeder and main track side-by-side. Which would be better and why? Matters like your sidetrack storage play into that debate, as well as the issue of bi-directional vs. unidirectional routing, climate controlled vs. open storage, etc. That all comes home to roost when I am trying to decide just how tight of a turning radius is tight enough,.. at what point this, or other design explorations, enter the realm of diminishing returns.
And so, I, Dan the Blogger, note and catalog conflicts and inefficiencies big and small!

Asko K. said...

Yes, it seems weird that existing solutions haven't bothered more on this. Our design has separate "storage stacks" for empty vehicles and separation of incoming and outgoing traffic at high-capacity (parallel) stations.

The best place for the storage stacks would be in between the incoming and the outgoing stations. Being able to react fast to demand spikes on the outgoing side is one of the main purposes of the storage stacks, and requires them to be nearby.

With regard to storage space, essentially any area will do. Parking halls especially well. This is also environmentally meaningful since there will be a surplus of parking hall area in the future, and other means to recycle such buildings are fairly limited.

Andrew F said...

Dan, I'm not saying that vehicle storage is irrelevant, I meant that when I thought about it, it seemed a straightforward problem to address. I could be wrong!

You're right that dense areas will be the trickiest for vehicle storage/buffering. But that is especially so in a 2D/planar paradigm like we have (for the most part) with cars. PRT makes 3D designs very easy, which to my mind makes vehicle storage/staging/buffering a fair bit easier, even in dense areas. In downtown Manhattan, you could just add sidings above the main guideway. For longer term storage (vehicles that are not needed for hours), it should not be a big problem to move them a few hundred meters or even a km or two to store them somewhere where space is not at a premium. All you need is a sufficient buffer to allow slightly more distant stores of vehicles to respond in the change in demand without causing undue delays.

Beyond that, how you have the local buffer for the station (or whether you have such a buffer) depends a lot on the station design. Parallel stations like ULTra would not seem to me to require an empty vehicle buffer until you get to very high capacity stations.

It is interesting to thing about, certainly, I guess it just seems odd to me to think about it in isolation from station design, as they seem highly related.

Dan said...

Hello Asko! I see you have improved your website since I last visited. I particularly like your video, which seems extremely well thought out. There are a so many arguments for PRT, it is exceedingly difficult to give each one the proper amount of emphasis, while putting them all in a logical order. You did as good a job as I have seen.

About those parking “halls” - (I take that to mean multi-level covered parking …usually in the form of a rectangular spiral…) Actually you have hit on a limitation of the system I am working on. These structures often have very low ceilings, and I need about 3 meters. Also the move from private offices to cubicles and bullpens means many office buildings have undersized garages since they are fitting more people in each building. These garages, when public, are getting pretty hefty parking fees in most downtown areas that I know of. Abandoned malls and strip centers are plentiful on this side of the Atlantic, though, and offer a great place for people to leave their car behind. I think taking the station indoors would be a great way to revitalize these spaces, since a station would generate a stream of pedestrians.

Andrew, I may be biased by the one city (Houston) which is easiest for me to consider, but when I walk around and look for sites for stations, I have a hard time finding room for anything of any size. I am isolating the two because there may be little room for a station in the first place. In “Uptown” Houston, for instance, they make a big deal out of lighting up the trees (and buildings) for Christmas (they even have a ceremony with fireworks) They are not going to cut these down for PRT! All of the surface parking is packed, and there is a huge waiting time to cross any street. With the number of bad weather days (heat or rain) we have to deal with, people would rather take their air conditioned cars than even walk very short distances, so a meaningful system needs to get people, more or less, door to door.
The best answer seems to be micro stations, which, of course, have little buffer capacity. These merchants, however, are sitting on the most expensive real estate in the city and have everything to lose. I can only assume any PRT system, badly needed as it is, would be HUGE battle on many fronts. Whatever storage there is will need to be shaded, and remote. Downtown Houston, which is 6 miles away, is better for PRT, except for the elevated crosswalks between buildings. Anyway, my observations have led me to believe that solutions that minimize horizontal area will be generally more acceptable. Lines of stored pods, bypass track around them, and long stations all make the system harder to incorporate and give the effected businesses more to complain about. We, in the US, are better at asserting our personal rights than sacrificing for the common good, I’m afraid.

Andrew F said...

Dan, you could store a lot of vehicles above the roof of a low-rise building (a big box store or mall, perhaps), perhaps even without having to make substantial structural modifications to the building.

It might indeed be tricky to retrofit existing buildings and neighbourhoods to maximize the utility of PRT. I suspect that value of having a PRT station integrated into the building (like the value of having free on-site parking) will be sufficient to encourage these modifications. Of course, I expect any new construction to have architectural features that would leverage PRT, such as a second lobby at the level of the PRT station--perhaps even the main lobby in terms of how the elevators and reception are set up in a building. I'm sure there is plenty a creative architect and structural engineer could do to integrate PRT infrastructure into existing buildings, but I don't think it will be as pretty or as useful as structures designed with it in mind.

Do you have an address for me to streetview to get a sense of what you mean? I don't think it should be a hard sell to displace some surface parking to put in PRT stations, not matter how in-demand those spaces are. Per unit of land coverage, a PRT station ought to be able to support more visiting patrons/employees/residents than surface parking lots.

Dan said...

If you want to take a virtual walk, try starting at Post Oak and Westheimer, or, a bit over a kilometer east is “Highland Village” which still has their palms festooned with Christmas d├ęcor in the Google “street view.” Note the shiny stainless steel ring over the intersections, as well as the polished arches over Post Oak as you go north past the Hilton. The Bus stops are upgraded.

The whole area on the south of Westheimer is “The Galleria,” which is a high-end mall and hotel complex favored by rich consumers from south of the border. Just south of that is the Williams Tower, the fourth largest building in Texas, at 65 floors, further adding to rush hour congestion. The street view pics must have been shot on a Sunday, because there is no traffic. Anyway, the point is that there are only a couple of meters from curb to private property… not enough for most station designs, and this is true for the whole area.

The properties themselves are basically being held as investments. Any low-rise structures on them are simply a means to draw a bit of income and pay the taxes in the meantime. Even though there may, physically, be enough room in areas now used for parking, there is no incentive for the owners to give up any sizable amount. A large station will never create a pedestrian rich area. It is just too hot. We just don’t walk around bathed in sweat, something that anything over 4 or 5 minutes of afternoon heat pretty much guaranties. This reminds me of the matter of the temperatures inside of stored PRT vehicles. In Houston, vehicles parked in the sun quickly get hot enough inside so that metal parts such as seatbelt catches burn your skin.

My take on it is to use micro stations, probably through a lease arrangement, since the property owners would never want to sell. They would be VERY worried about the effect on the “look and feel” of the area, and it would be very difficult to get any kind of a consensus. Again, because of the heat, stations would not have a widely felt effect on foot traffic, so merchants with no station would have nothing to gain. All businesses would welcome a non-stop shuttle to the Galleria, though, so long as it was discrete in appearance. Given a choice, I would bet most merchants would prefer to have an in-store “portal” to a transit system that was unseen…(rooftops? back alleys?)

BTW, a system like the one I advocate would easily allow stations in any commercial space, on the first floor as well as the second. Most rooftops here wouldn’t be rated for the weight of PRT. We have no epic snow events, after all.

ItsEric said...

A simple solution is to build bays above the parking lots with carport structures to shade them. Three side by side tracks that hold 50 cars would need roughly 20 ft by 600 ft (assuming cars are 12 by 6 ft). Two tracks could then be turned off until those cars were needed. Long side tracks minimize switches, cost, & complexity. Finally, backing the cars in allows for quick response.