Marginal Station vs. Marginal Length Cost

(Note: It’s been a while since I’ve posted on The Gondola Project. It’s been a busy summer with lots of changes to our company and our site. We’ll let you all know about the details in the coming months, but in the meantime, I’d like to extend a big thanks to Nick and Charlotte for holding down the fort while I’ve been awol.)

One of the problems the cable industry faces (like most transit industries), is prospective customers who fear nuance. Prospects often don’t care about the complexity of a system, they simply want to know how much it costs “per kilometre” (or “per mile” for our American friends).

Here’s the problem: It is virtually impossible to provide a per kilometre cost for a cable transit system. In fact, it’s virtually impossible to provide a per kilometre cost for any transit system, period. It’s kind of like that old idiom—how long is a piece of string?

No where does this become more obvious than with the relationship between the length of a cable transit line and  the number of stations within a cable transit line.

Let’s assume, for example, a given 1 kilometre long cable transit system that has two stations and costs $8mm. Let’s call it Line A.

Now let’s assume a second cable transit system. This one is the same length of Line A (1 kilometre) but has a total of three stations rather than two with all else equal. Let’s call this system Line B.

Now, let’s assume a final third cable transit system. This one has only two stations but is double the length of Line A—it’s 2 kilometres long with all else equal. Let’s call this system Line C.

Okay? Got it? No? Let’s review then:

  • Line A: 1 km, 2 stations.
  • Line B: 1 km, 3 stations.
  • Line C: 2 km, 2 stations.

Which line is more expensive? Line B or Line C?

People who imagine the problem as a question of cost-per-kilometre will invariably say Line C is more expensive than Line B because Line C  is double the length of Line B.

Problem is, they’d be completely, 100% wrong.

In cable, the marginal cost of stations is almost always more expensive than the marginal cost of length.

People considering a cable transit system of their own need to understand that. Per-kilometre costs estimates are blunt tools that don’t tell you what you really want to know—and they often lead to early-stage financial estimates that tell a completely false story.




What Can We Learn From The Pakistan Cable Car Accident?

Vernacular, improvised ropeways in places like Pakistan, India and Nepal are common. They are often over-crowded, poorly built and maintained and too-often are involved in accidents.

As we’ve demonstrated before, cable cars, ropeways and gondolas are amongst the safest transit technologies in the world.

But that doesn’t mean accidents don’t happen.

A few days ago in Pakistan a cable car system snapped, sending 8 passengers plunging into the swollen Indus river below. No survivors were reported.

A similar incident occurred last year in Nepal.

But before everyone jumps to conclusions and assumes this is typical of the technology, please remember that both of these systems were improvised, vernacular installations. These human-powered systems tend to be hand-built by locals lacking the proper resources to build effective installations. Incidents such as these are often characterized by low-capacity vehicles overloaded with passengers straining the system’s upper limits past its breaking point.

Yes cable transit systems are safe – but only when designed, built and installed by qualified, knowledgeable professionals with access to the proper tools and resources. That stands for virtually any product.

The reality is this: These types of systems exist in plenty throughout the developing world. In regions where topography reigns supreme and money is scarce, these improvised are going to flourish. In many ways, that’s a good thing. Here are people using their own know-how and limited funds to solve very clear mobility challenges. We can’t fault them for improving their lot.

What we can do, however, is assist them in their endeavours. So while no one doubts the tragedy of these incidents, they do point to clear opportunities for the cable industry to both do some good and increase their market presence:

1. Most major corporations make donations to some charity or another. How about an in-kind donation of a slimmed down, professionally-designed system that can stand up to the rigours of these environments?

2. Development money in these parts of the world is a-plenty. Why not create a low-cost ropeway system specifically designed for this market?

3. A combo approach of both one and two: How about partnering with local development agencies and offer in-kind donations of ropeway services to check on these various installations, report on their safety and rehabilitate them where necessary?

4. How about setting up ropeway engineering schools and programs in these isolated areas to educate the locals about proper techniques and providing the necessary resources to realize those goals? Or take it one step further and build small branch-offices to do just that but within an existing corporate structure?

As the cable industry becomes more-and-more a player in the city building and urban transport markets, corporate social responsibility (CSR) will necessarily become a key part of their marketing strategy. And that’s not going to be a choice.

The major ropeway manufacturers will be compelled by market forces to engage with communities such as these in unique and innovative ways simply because the current city building industry privileges those companies that do. City building, whether we wish to admit it or not, is a pay-to-play game.

So rather than run some bland, run-of-the mill CSR endeavour, how about spotting opportunities to get a little dirty and actually do some good?

This is clearly one such opportunity.



What Can The San Francisco Cable Cars Teach London?


Image via Wikipedia.

If nothing else, the London Cable Car raises an interesting question:

When should a transit line be fully fare-integrated into a transit network and when should it not. For those unfamiliar, an additional fare is required for people to ride the London Cable Car despite it, ostensibly, being a part of the overall Transport for London network.

But should that be the case?

We have no shortage of examples of transit systems who operate on a similar model:

In Toronto, to access the islands, a ride on the ferries cost an additional fee above and beyond a standard transit ticket.

In Portland, a ride on the Aerial Tram costs an additional fare – with the exception of students and staff of OHSU. OHSU is the primary beneficiary of the Aerial Tram and riders affiliated with the operation ride for free.

In Medellin, Linea L services a nature preserve and requires an additional fee beyond a standard transit ticket.

Users of Vancouver’s Skytrain must pay an additional fare if they wish to use it to travel into the city from Vancouver International Airport.

Seattle’s extensive network of ferries all require separate fees despite being an integral component of transport in the region.

So what’s the difference? Why do people not have issues with the examples above but do with the London Cable Car?

The difference, I suspect, is mostly about the target market. The above examples largely service: a) a small subsection of a local population or b) a large local population who is likely to use the service very irregularly.

The system the London Cable Car represents most closely, coincidentally enough, is the San Francisco Cable Car. Both systems operate in central, rather than peripheral areas. They both feel as though they should be intertwined into their respective networks. Furthermore, the San Francisco Cable Cars are priced outside the normal San Francisco transit fare. Prices are five bucks for a one-way trip or thirteen bucks for a day pass.

Like the London Cable Car, those prices are well outside the budget of any standard commuter.

Interestingly though, the San Francisco Cable Car is used by regular commuters – they just don’t pay for it.

In 2007 auditors found that 40% of riders on the system were not paying fares. Now there’s no way to confirm whether it’s locals or tourists who are breaking the law, but assuming it’s locals is not an unreasonable assumption as those are the people who are most likely to understand how to catch free ride successfully.

People such as Jarrett Walker have argued in the past that the San Francisco Cable Cars add little to no value to San Francisco’s transit network, but stats like this suggest otherwise. From my understanding, it’s very common practice for locals to hop-on and hop-off the cars for quick trips around town without paying fares – and for the fare collectors to turn a blind eye to it.

Which in turn implies that the high fares paid by tourists help to subsidize the system for local commuters. It may be illegal, but it’s an interesting reality.

Now in no way, shape, form or description am I suggesting that fare evasion is right and justified. But I don’t think you’re going to find many San Franciscans who have a problem with it. In reality, it’s little different than the Swiss practice of selling ID cards to locals that grant them 50% off all transit in the country. The only difference is the degree of formality.

Londoners won’t be hopping on and off the Thames Cable Car any time soon, but San Francisco does provide an interesting lesson. If Londoners were allowed “free ride” on the backs of tourist fares, I suspect few people – if anyone – would’ve objected to the system.

In fact, were that the case, locals might even come to love it.



Dutch OV-chipkaart. Does it make transit more or less equitable?

A bus-based OV-chipkaart reader in the Netherlands promises to make transportation easier for millions of Dutch strap-hangers. But is it an equitable solution?

A Thought Experiment: You live downtown and you’ve got to travel a distance of 1.5 km, drop off a package and return home. If you lived in Toronto (as I do), the trip would cost you $6.00 round trip by subway.

Would you pay it?

A great many people would probably say answer ‘no’ to the above question – I know I wouldn’t. Six bucks, after all, is a lot of money for what is nothing more than a 40 minute trip.

Most public transport fare schemes are tilted in favour of the long-distance commuter. All-in or even zone-based travel means the further you travel, the more you save. Short-distance commuters, on the other hand, are often left with the choice of either walking, biking or paying high prices for little distances travelled.

But a new Dutch transit fare scheme seeks to change that.

With little fanfare from the rest of the world, the Netherlands has been slowly moving towards a unique system of integrating all public transport options around a single national fare card and pricing scheme.

That system, the OV-chipkaart, has admittedly met with its share of bumps along its way towards full implementation, but the single-trip pricing scheme is what should interest people here most.

Generally speaking there are three dominant fare structures within public transportation:

  • Zone-based fare structures (such as in Hong Kong or London) set a specific price to travel within zones and between zones. For example: If you originate in Zone A and terminate in Zone A, you’ll pay $1.00. If you originate in Zone B and terminate in Zone A, you’ll pay $2.00.
  • Unlimited one-way fare structures (such as in New York City or Toronto) set a one-way price to travel from Point A to Point B using an unlimited number of transfers and vehicles with stop-overs typically not allowed.
  • Time-based fare structures (such as in Rome) allow a rider to travel to an unlimited number of destinations, using an unlimited number of transfers and vehicles within a given time window. In Rome’s situation, for example, one pays €1.00 for 75 minutes of unlimited travel time anywhere within the network.

Granted, all of the above are subject to things like monthly passes, day passes, etc., but for the sake of simplicity, we’re talking only about single-trip fares.

The OV-chipkaart is a hybrid of all three. The system works like this:

Every year a base rate per trip is set nationally for all public transport agencies. In 2012, that base rate is €0.83.

Each transport agency within its own defined area is then allowed to set their own per kilometre fare. In the case of Amsterdam’s GVB, for example, the per km fare is €0.142.

Traveller’s must use their OV-chipkaart to “check-in” using an electronic reader or turnstile when they begin their journey, make transfers between vehicles and “check-out” when they finish their journey. The OV-chipkaart calculates the fare and deducts it from the user’s electronic “purse.”

Interestingly, riders are allowed a 35 minute window between transfers. That is, if a user “checks-out” of one transit vehicle but then boards another within 35 minutes, the base rate is waived and the journey is considered ongoing.

That allows for a huge number of errands and stop-overs to occur that most other systems would strictly forbid.

Ultimately, what this system does is drive up transport prices for long-distance commuters while driving down the price for short-distance travellers.

Applying Amsterdam’s fare structure, extrapolating it to a place like Toronto and using the 1.5 km example given earlier, the total round-trip would cost ~ $1.60, not $6.00.

Were there to be a stopover longer than 35 minutes (say, for a work day), the price would still only be ~ $2.60

Conversely, someone travelling 20 km to work would see their round-trip price increase from $6.00 to ~ $9.25.

Current practice in transport planning does tend to disincentive public transportation for those that live, work and play within central, geographically small areas. Paradoxically, these urban dwellers are the most likely to ride public transport but are penalized with the highest per km fares.

On the flip side, those long-distance commuters within inner-suburban and suburban areas are the least likely to use public transportation and any increase in fares will only exacerbate the situation. Furthermore, large swaths of inner suburban areas are dominated by the economically disadvantaged who are already living paycheque-to-paycheque. A more than 50% increase in their public transportation costs would be devastating.

Our basic understanding of economics tells us two contradictory things here: One the one hand, we understand the idea that people should pay a proportional amount for a specific good or service. But at the same time, we also understand that the more of a good or service you buy, the less you should have to pay per unit.

So the question is easy to state, but difficult to answer: Which fare structure is more fair?



Thought Experiment: The Cost of Multi-modality

Whether we’re talking about Light Rail, Urban Gondolas, Monorails or PRT systems, I’ve recently begun to hear an argument against multimodal transit systems that I’ve yet to hear before: The cost.

From a transit agency accountant or city finance perspective, multi-modality is nice in principle but incredibly costly to implement.

Let’ s walk through this issue with a simple thought experiment:

Imagine you’re in charge of a 30 year old transit agency in a medium-density, medium-sized American city that’s worked exclusively with a fleet of ~200 buses over those decades. Recent population growth and an infusion of money from the Feds allows you to expand both your service level and coverage.

You have two choices:

Your first choice is to stick with what you know and buy another ~ 50 buses. The buses option isn’t ideal. The curvilinear layout of your road pattern means buses routes will be highly circuitous and indirect, thereby increasing travel times and headways between vehicles.

On the flip side, you’ll only have to hire a handful more mechanics and make a relatively minor expansion to your maintenance and storage yards. You’ve already got all the equipment you need; and well-functioning procedures and processes are already in place. In other words: There’s no learning curve.

The cost savings associated with no learning curve mean you can expand the system to a relatively great degree. But the bus compromise mentioned above suggests your expansion plans will generate only marginally more ridership.

Your second option is to go with a different technology. For the purposes of this thought experiment, it doesn’t matter which. You could go with Light Rail, Gondolas, the CableRailGyroCopter, the Dutch Superbus, it doesn’t matter.

What does matter is that whatever your technology choice, it will provide a superior level of service to the buses you’ve been running for the past generation. The route alignment will be direct and will likely cause a decent increase in ridership.

The catch is that your agency has no history of dealing with any technology other than bus. That means there’s a learning curve – a steep one.

You need new maintenance facilities, new processes, new drivers, new mechanics, new schedules, new everything. Not to mention all the consultants you’ll have to pay to understand and realize this brave new transit world. After all that, there’s not much money left over to actually build the transit system. That means you’ll only be able to build half as many kilometres of new transit lines than if you went with the simpler bus option.

That situation is further reinforced by the lack of economies of scale you will experience in Operations & Maintenance that will inevitably drive up your per trip cost per rider. Were you to go with bus, your per trip cost per rider would actually decrease due to the economies of scale you would experience upon opting for the first choice.

So which do you opt for?

I currently know of no research that delves into this issue of the cost of multi-modality in public transit, but think it to be a fascinating area that’s worth exploring. If you know of any such research or have any personal/professional experience in this area, please give us your thoughts in the comments below.



Should It Be?

A strange thing happened during yesterday’s discussion of the spiralling costs of the London Thames Cable Car:

Like any time before when the topic of urban versus resort installations comes up, the de facto response is: Well, duh, of course it’s more expensive to build in cities and for government.

You’ve said it. I’ve said it. We’ve all said it. We’ll probably be saying it tomorrow.

But the question is this: Should it be?

We have no shortage of examples of tourist-oriented systems built in urban areas. Those systems are always more reasonably-priced than their transit-oriented brethren. And yet they both share the common variable of still being located in urban areas.

Why does the addition of a layer or two of bureaucracy justify an orders-of-magnitude extremely excessive cost increase for a product whose only real difference is said layer or two of bureaucracy?

Maybe that’s an incredibly naive question to ask. Maybe. But it’s also a question that needs to be asked.



Exploring the Thames Cable Car Costs

Over the weekend it was announced that the estimated project cost for London’s Thames Cable Car (Gondola) has ballooned to an estimated £60m. For those interested, that means the system will cost roughly $100m USD per kilometer.

With the possible exception of the Caracas Metrocable (whose finances are discussed here), the London Thames Cable Car will easily be the most expensive gondola/cable car ever built. It’s even more expensive than the overpriced Burnaby Mountain Gondola, whose cost has also yet to be explained or justified.

As we discuss here, the London Thames Cable Car appears to be nothing more than the latest example of largely English-speaking transit agencies’ unwillingness and/or inability to reign in costs related to transit projects.

Whether this is an example of scope creep, pork barrelling, corruption or just pure out-and-out incompetence is virtually irrelevant at this point because all or some clearly have a role to play in this debacle.

As someone who happens to know a little bit about cable transit systems, let be me completely blunt: There is absolutely, positively, completely no reason whatsoever this project should cost London taxpayers ~$100m USD. Not a single good reason:


ONE. Off-the shelf MDG technology is being used.

While we have no confirmation of this fact, we can use a little something called logic to figure it out. Construction on the project started just a few months ago. As the goal is to complete the system by next summer, the only possible way in which to do that is to use MDG technology.

3S technology would be the sexier (possibly even better) choice here, but the reality is this: MDG technology has virtually off-the-shelf availability; 3S has to be built to order. That’s why an MDG can be turned around in such a short period of time.

Widely available renderings also indicate MDG technology.

All-in, an MDG system can be built for $10m – $30m USD per kilometre. Max.

At their most expensive, Medellin’s Metrocable systems were coming in at ~$25m USD and that included intermediary stations, turns, 4,000 pphpd capacity (compared to London’s 2,500 pphpd), land acquisition and all station and tower architecture.


TWO. Where the technology is manufactured invalidates questions of where it’s built.

Okay, sure. Medellin isn’t London and it certainly doesn’t cost as much as London. But remember: As the majority of the cable system itself is manufactured in a western-European location (France, Austria, Switzerland and/or Italy), that means the cost of the systems’ electro-mechanical components (cabins, cable, towers, stations, etc.) will not vary much from place-to-place.

The only thing that’s likely to cause any sort of shift in price is currency and/or commodity fluctuations. But as the Euro has been depreciating against the British Pound for much of the last two years, shouldn’t the price actually be decreasing?

The one counter-argument to this could be if much of the system is being manufactured in Switzerland – in which case the rapid over-inflation of the Franc could be leading to these increases but a) much of that over-inflation has been recently stemmed due by Switzerland’s Central Bank and b) as we understand it the system is being built by Doppelmayr not Garaventa.

This is important because Doppelmayr is an Austrian company that trades in Euros and Garaventa is a Swiss subsidiary that trades in Francs. In other words, somewhere between 75 and 90% of the cost of this system is being incurred purely in London alone.


THREE. The capacity of this system invalidates the need for large scale station infrastructure.

As reported, the Thames Cable Car is expected to move roughly 1,000,000 people in its first year of operations with a throughput of 2,500 persons per hour per direction. One million looks like a big number, but it’s really not when you consider how many hours there are in a day and how many days there are in the year.

(Note: The Londonist riffs on this concept with a deplorable calculation that demonstrates how only 228 people will use the system per hour. This calculation is a gross over-simplification of the problem because it spreads ridership evenly across every day and every hour of a 365 day year. Of course we know that ridership has peaks and valleys from hour-to-hour and day-to-day especially when you inject a massive peak such as will occur during the Summer Olympics. This calculation is therefore nothing more than brazen misinformation and irresponsible commentary.)

Oversized station architecture typically accounts for the bulk of costs in a cable system such as this, but given the modest number of people this system is anticipated to move, there is absolutely no reason to invest in large scale stations.

To demonstrate: The Koblenz Rheinseilbahn utilizes the above-mentioned 3S technology and moves ~3,600 pphpd. It has been an enormous success and I’m told is moving tens of thousands of people per day due to the bi-annual (and inexplicably popular) BUGA horticultural festival.

The Koblenz Rheinseilbahn is also only 1 km long and cost ~$20m USD all in.

This is what the Koblenz stations look like:

That's the entire station, infrastructure and all. Image by Steven Dale.

Now compare that to London:

See the difference?

Granted the London system has two things that the Koblenz system does not. Owing to Koblenz’s current status as a temporary installation, it does not have the maintenance bay and custom-designed towers that the Thames Cable Car will. Fine. But do those two items justify the Thames Cable Car’s absurd price premium over the Rheinseilbahn?

Not when the Rheinseilbahn carries almost double the number of people.

Transport for London and Mayor Boris Johnson owe the people of London an explanation – particularly as they now plan to pay for it “out of the rail budget.”

Suffice it to say, this isn’t going to win Urban Gondolas any fans – likely just a few more enemies in one of the most highly visible cities in the world.


Related Posts Plugin for WordPress, Blogger...