PRT service vs. Current Express Bus System

This will begin to answer the question,
"How might PRT compare with Bus Rapid Transit?"






Excerpted from correspondence between Get On Board! PRT and the group Puget Sound Regional Monorail.

Feb. 12, 2003: ...a set of rough calculations for demand on a PRT starter system that works out quite nicely--

Coverage Area (ride origination):

North Seattle & Shoreline [express bus] park & ride lots
County data:
Number of lots: 18
Total spaces: 3365.
High usage (90% or more of spaces full end of morning rush) at 7 lots.

Question: Can PRT handle this level of demand?

System assumptions:
PRT station at each lot; 3 berths per station (18x3=54 berths);destinations primarily Downtown and U District, so stations will be needed in thoseareas. BUT we're concerned in this example with knowing whether a PRT system can launch enough trips to handle present P&R demand. Anyways, if it canlaunch them it can receive them.

Demand assumptions:
Let's make it tough on PRT by assuming 95% of spaces full at ALL lots.

Each parking space used = 1.2 people.
3836 people, or 1279 per hour.
All demand between 6 and 9 AM (again making it tougher by cramming demandinto 3 hours instead of all morning)
PRT service assumption: 1.2 persons per vehicle (same as automobileassumption)

Results:

1065.6 launches per hour
17.75 launches per minute
0.986 launches per minute [1 every 60.8 sec.] per station
0.328 launches per berth per minute [1 every 3 min 4 sec] per station

This is only, maybe, 15-25% of the maximum throughput for this manystations/berths.

Minimum requirements for downstream stations (in Downtown and U District, must be capable of receiving vehicles launched above):
      If we assume 4arrivals per minute (240 per hour) as the rough capacity per berth, 4.4 berths would be needed to handle 1066 vehicles per hour. In other words,MATHEMATICALLY you could get by with only two 3-berth stations, one Downtown andone in the U District. But in reality we would want [for reasons of wider access and convenience] maybe 6 stationsdowntown and 4 or 5 in the U District, along with stations at places likeEastlake and Cascade/REI. Thus we would probably avoid most station congestion("wave-offs" if the station is full) that might be caused by vehiclesarriving in big clumps. With at least 3 berths per station we could expect thepace at stations would not be too hectic.

Required PRT fleet size options, assuming--

3 trips/vehicle/hr: 1066 vehicles
4 trips/vehicle/hr: 800 vehicles
5 trips/vehicle/hr: 640 vehicles
6 trips/vehicle/hr: 533 vehicles

Just a guess, but there would probably be minimal deadheading of emptyvehicles. For instance, vehicles arriving downtown from the north end wouldthen pick up people going to the U District or Northgate. Vehicles wouldthen be distributed to the park & rides as needed to make more runs todowntown.

--DSG




March 27, 2003:Conceptually, one must remember that, at the start, PRT is likely going to be serving not an entire corridor, but a section of a corridor and theareas on both sides. For instance, in the Everett to Downtown Seattlecorridor, an initial PRT segment might cover the area roughly between Northgateand the U District. The PRT service area would be a N-S-E-W area straddlingthe corridor, so at that point you have to get out of line-haul thinkingand think in terms of serving origins and destinations WITHIN the area. Idon't think I need go into a proof of PRT's capability in such alocal/crosstown context. People from outside the area would be able to drive to a P&Rand take PRT to Northgate or Roosevelt or the U, but I think most servicewould be residents within the area.

Note that I'm still assuming loops, not a linear system. Linear meanscomparing PRT to conventional transit using Passengers Per Hour Per Direction, and that is one of the weapons used by planners and consultants todisqualify PRT. Even when Ed Anderson talks about Taxi 2000 operating in acorridor alignment, he's not talking about a two-way line, he's talking about aseries of loops, roughly a half-mile on a side, that are laid out in a rowinstead of a grid, and a freeway runs down the middle of the row.

Assume we have 4-5 PRT loops in the initial Northgate-U District segment,and let's call that 12 miles of guideway with 480 vehicles. The fleetcould provide, roughly, 3000 rides per hour to about 3600 people. This isequal to 28% of the average hourly transit ridership for all of King County.

After the first phase is up and running, we'd then add another network ofthe same general size, let's say 4-5 loops that would connect to the GreenLine and cover Greenwood up to Shoreline. Adding another 480 vehicles, the fleet now serves 7200 people per hour. After that, a third phase might be U District to Downtown, Capitol Hill and S. Lake Union-- maybe we'd add750 vehicles in this expansion, since now we'd be able to offerShoreline-Downtown corridor service as well as local/crosstown service. The fleet would now be 1710 vehicles, able to offer roughly 10,260 rides per hour. Thequestion now is whether this is enough capacity.

Follow me now as I distill the comparison below--

I've already established... 1000vehicles could theoretically handle (more than easily) the number of riders(about 1100 launches per hour between 6 and 9 AM, about 3900 people) comingfrom N. Seattle P&R lots. Because I based that exercise upon the number ofparking stalls, it doesn't matter whether people are getting on a Metro busor an ST Express bus. But let's make it hard for PRT. Let's assume thereis no overlap, let's say the 1000 vehicles are handling ONLY Metro riders. If ST Express service ceased, how many additional people would the other710* PRT vehicles (hourly capacity about 4,260) have to serve in theShoreline-Downtown corridor?

According to ST's own Ridership Report, the Everett-Seattle routes had173,000 boardings in the 4th quarter of 2002. They also say theEverett-Northgate segment had 9,400 boardings. Therefore, the Shoreline-Downtownsegment was about 163,600 boardings in that quarter. (Until we build PRT toEverett, the 9400 people would continue to take the bus).

There are 91 days in a quarter, 65 of them weekdays. The four ST routes(510, 511, 512 and 513) make a total of 79 daily runs M-F (40 inbound, 39outbound), 33 on Saturday and 30 on Sunday.

Let's now estimate the demand on ST Express during an average peak hour.

86% of service is weekday.
86% of 173,000 is 148,780 boardings during the 65 weekdays. The averageis 2,288 per weekday.
14% of 173,000 is 24,220 boardings on the 26 weekend days, or 932 perweekend day.

Rule of thumb: 10% of daily rides come during 1 peak hour. Thus, peakhour transit rides in King County is about 313,000x0.10=31,300. So onweekdays, peak hour boardings on ST routes 510-513 is about 2288x0.10=229. Thisis 2% of the hourly capacity of the entire 1710-vehicle [PRT] fleet describedabove, and 5% of the hourly capacity of the 710 [PRT] vehicles at the * above. I was concerned I had made a math error. But I count 4 ST runs from Everett toSeattle during 6-7:00 AM. 229/4=57 boardings per run, so I think myestimates are reasonable.

Finally, we conclude by taking the 229 ST Express boardings and add the1100 Park & Ride PRT launches. The total is 1329 for one hour, or 13% of the PRT capacity. Thus, there is plenty of PRT capacity left over to providea higher level of Everett-Downtown travel, other trips within the network,and operational considerations like empty vehicle movement.

--DSG

Get on board! Personal RapidTransit