I had some free time yesterday afternoon (12/19/2006), so I visited both endpoints of the tram to check it out in operation. (I am not an OHSU employee, so I did not try to board.)

The first thing I noticed was just how quiet the Tram actually is… you could not hear the cables or pulleys operating over the noise of other construction going on in the area. After construction winds down, it will be interesting to hear if the tram is audible above ambient traffic noise.

I watched the upper landing from about 12:00-12:30 and the lower landing from about 1:15 to 1:30.

A tram was departing every 5 minutes, and I counted no fewer than 10 people boarding each departure (or deboarding each arrival), with the largest crowd being 18 people. The tram is open to just OHSU employees at this time, and I imagine that a good number of these boardings were folks checking out the tram on their lunch hour, but still, the ridership is encouraging. Given a median boarding of 14 persons at each terminus, departing every 5 minutes, that means that ridership was at a rate of 336 per hour during the time that I observed the tram. The official initial forecast of 1,500+ per day seems plausible.

At the lower platform, I chatted with one of the tram operators for awhile… she certainly loves her job and was happy to answer questions. (Unfortunately I have forgotten her name.)

We talked about the recent windstorm and she said that she was working that day and that the peak winds experienced by the tram were 45mph; she barely felt it on-board, however some of the passengers who were not used to the tram did reach for the railings during heavy gusts. She said that she didn’t really get a sense of just how windy it was until her break time when she would step off at ground level and feel the wind outside.

There are two operators (one per tram car), and a third staff person (a shift supervisor?) rotates through operator duties so that the main operators can get breaks, lunches, etc. There are two full shifts per day.

Over on the Commissioner Sam blog a few days ago, a debate opened up about ventilation, air conditioning, etc. I ask the operator about this and she told me that there are multiple fans in each tram car, but no air conditioning. I asked about people being stuck in the cars should the backup drive systems fail, and she said that even when the drive systems and main power are down, there is a backup generator at the lower landing which provides power to the cabins for lights, controls, fans, etc. Whether the fans will be sufficient in hot weather is still up for debate, at least until summer I guess.

The operator told me that they are still running a bit slow as they ramp up to full operations, with the complete trip now taking over 3 minutes, and they will soon ramp up speed to achieve the normal operating time of 2min, 40sec.

After the flip, my “not packed in like sardines” arguments and “cost per passenger mile” estimates…

**Numbers, numbers, numbers…**

Over on the CommissionerSam blog, I made some estimates about capacity, cost per ride and cost per passenger mile. In the hope that those estimates will be of interest to PortlandTransport readers, I am reposting them here:

First, the complaint has been leveled by some folks that tram passengers will be packed in “like sardines” and will expire from heat combined with cramped conditions…

**Not Sardines**

The interior floor of a tram cabin is over 200 sq. ft, with a published maximum passenger capacity of 78 persons, plus the operator, for a total of 79 people.

At maximum load, that leaves about 2.5 sq. ft. per person. The Portland Streetcar’s published maximum load puts it at about 2.4 sq. ft. per person.

However, given a typical full load for a MAX train or bus (which have more seats per sq. ft. than the streetcar or the tram), 3.5 sq. ft per person is more typical.

If we apply the 3.5 sq. ft. minimum space per person to the tram cabin, we get 56 passengers plus the operator.

Now, let’s assume for a moment that people won’t want to ride the tram if they feel like “sardines”, and that many people will want to cluster around the windows, etc., making some areas tighter and some looser. Let’s use a generous 5 sq. ft. minimum space per person instead. That would give us comfortable room for about 40 people.

The tram is physically capable of making a journey every 5 minutes. Let’s assume though that sometimes things don’t operate efficiently, people take longer to board than predicted, etc., and that the trams depart every 10 minutes. *(UPDATE: Based on my direct observations on 12/19/2006, Tram passengers have no problem with quickly boarding/deboarding and the trams consistently depart on time, but I’ve left my original 10 minute figure in place here to prove a point.)*

Let us further assume that given all the artificial constraints and concessions I’ve given tram performance here, that “capacity” crowds of 40 per car only show up during 4 peak hours and that during the remaining 12 or so hours only 10 per car show up.

Given all of the artificial limitations and concessions that I’ve made here to avoid packing people in like “sardines” (where’s the oil?), let’s add up the capacities:

4 peak hours * 6 trips * 40 persons per direction = 960 persons per direction at peak hours (240/hour).

12 off-peak hours * 6 trips * 10 persons per direction = 720 persons per direction (60/hour).

That adds up to a daily ridership of 3,360 (1,680 per direction.)

Well, guess what, predicted ridership for the tram is only 1,500/day initially, ramping up to 5,500/day by 2030.

So, even assuming A) trams run at only half their design frequency and B) people only allow themselves to fill a car to half of the maximum capacity (now we’re operating at 1/4 capacity), the tram STILL has a capacity of about 3X the opening year ridership prediction, and comes well within the ballpark of the predicted ridership decades from now.

To really get “sardine” status at peak hour, operating at design maximums, you’d have to move 7,488 people over 4 peak hours. If the tram is ever _that_ popular, people won’t be cursing it, they’ll be clamouring for another one to be built.

(And no, I don’t think it will ever be that popular. The point is that the only way to really get “sardines” is if the tram is wildly over-successful.)

*(The following has been edited a bit since the original posting, but the numbers are the same)*

**Tram Costs**

JK asked: *Care to do cost per passenger mile?*

In a staff post on CommissionerSam.com, it was stated that “In today’s dollars the annual life cycle cost of the tram is estimated at $2,735,200 of which the City’s share is $409,280.”

I am going to assume a median ridership somewhere between 1,500 (opening day) and 5,500 (2030), say about 4,000. Note that the tram lifecycle is supposedly 50 years, so if they do hit 5,500 by 2030 and continue to hold that another 25+ years, the median will be much higher and therefore the cost per passenger mile much lower.

The tram travels 3,300 feet, which is .625 miles.

Assuming the ridership at 4,000, daily and weekend, that’s 1,460,000 annual rides.

Divide an annual cost of $2,735,200 by 1,460,000 annual rides, and you get a cost per ride of $1.87 per ride. Divide that again by .625 miles and your cost per passenger mile is: $2.99.

But is that a fair comparison? The tram creates a new, much shorter route between Point A and Point B. Should the cost be expressed in terms of tram passenger-miles, or in terms of comparison to the alternative, which would be buses weaving indirectly up the hill?

Google Maps gives directions from “SW Moody Ave & SW Gibbs St.” to “SW Sam Jackson Park Rd & SW Campus Dr” as a distance of 1.9 miles.

The tram is replacing a surface route of 1.9 miles with an aerial route of .625 miles. If you calculate the cost per passenger mile with 1.9 miles, you get a result of 98 cents per passenger mile.

The tram journey will take 2 minutes, 40 seconds. A journey by shuttle bus in traffic, with about 10 traffic lights and several major intersections, would take considerably longer, perhaps 15 minutes. The time savings has to be worth something, too.