The piano: Imagine

November 15, 2015

A guy, hauling a piano, on its bike in front of the Bataclan, the day after Paris attack


It’s to play “imagine” by John Lennon.

(this song has also been played on the piano, at the vigil this night in Vancouver, but the piano didn’t came by bike)

We refer here to the Canada’s Ecofiscal Commission report on congestion pricing, which received very large coverage in the medias, but also on Pricetags and the Stephen Rees blog.

We did a post series in 2011 on congestion pricing applied to Vancouver:

Frankly, I have failed to see a more comprehensive study since our 2011 series, putting hard number, and making “hard choice”. I am afraid the Ecofiscal report misses the mark too, however it is getting lot of media air, what can’t be a bad thing for advancing the congestion pricing case. We are going in a detailed analysis below:

The good

The report recommends a cordon pricing (in the form of a tolling of all water crossing in the region), something we have also recommended:

    Considering the Vancouver topography and the choke points responsible for most of the lower mainland congestion, the equivalent of a “cordon pricing” on the bridges seems the natural way to go.

Such scheme seems to raise many opposition we had already noticed in 2011. The mood is to not oppose frontally to the road pricing idea, but it is to rationalize the inaction toward it:

    Anything other that the “ideal” road pricing is not “fair” to the motorists

Those critics could be right, assuming the generalized economic gain largely offset the implementation and operating cost of the proposed model. Considering the overwhelming complexity of implementation of such a model, and the limited amount of congestion in the lower mainland, which limit the potential revenues, it is probably wrong headed:

    The model could be fair to the individual motorists, but could be unfair to the general interest: It could cost more to operate than the general economic gain it allows.
Some people are against road pricing, inferring it is unfair on the poor. - same could be said of pay parking, transit fare...

Some people are against road pricing, inferring it is unfair on the poor. – same could be said of pay parking, transit fare…

Of course, a cordon pricing is a cost/benefit trade off which is not perfect- driving from Langley to Ladner could be exempt of toll, when short haul using the Pattullo bridge could be… but the “choke point” are on the bridge not on the Highway 10… Of course the very idea of road pricing involve that poor people could also pay…but all the argument raised against also apply to:

  • Pay parking or gas tax
  • Transit fare, especially the zone system with its arbitrary boundary making the trip across the Fraser twice more expensive that the long trip from Langley to Ladner…

…and like the transit fare structure, the type of congestion pricing structure has not to be seen as permanent, but is a trade-off which can be reviewed in function of the technology progress.

The inescapable reality is that the road space is a scare resource, and so far the best known way to manage a scare resource is to put a price on it to align the demand with the offer. This can have some social and fairness implication: let’s address them, instead to use them as an excuse to do nothing.

The bad

The Ecofiscal report seems essentially geared for media consumption. Beside giving some air to the idea, what is already a lot, the report doesn’t seem as well researched as one could have expected, and it doesn’t bring any new element to the discussion. Data from [1] are mainly used afterward to make our points.

What is congestion?

Some, in their hostility to road pricing, don’t hesitate to argue, that road congestion is a good thing: doesn’t are vibrant and economically prosperous cities, crowded places? assuming that crowding is another word for congestion

It is not. here we take the pedestrian paradigm to illustrate the difference, since usually crowded pedestrian places are considered as a positive quality

pedestrian congestion in New York (left): people can't move freely and smoothly, traffic is typically stop and go and unpredictable: that is Level of Service F- heavy pedestrian crowding in Istanbul (right): it is dense traffic imposing slow speed, but still smooth and predictable moving: that is Level of Service E. Credit photos (2) and (3)

In the above examples: one could consider that the Istanbul picture is a desirable outcome for a place, the case is more difficult to be made with the New York picture. Transposing the above paradigm to the road traffic:

  • Busy roads are a desirable outcome, since it is a sign that the urban land space use is maximized
  • roads congested to a level where the traffic become not smooth and predictable is undesirable

Finding the desirable sweet spot is an exercise in itself, but generally speaking, people will agree that the congestion at the approach of the tunnel or the Pattullo bridge has well past its point of desirability. However we could have wished from the report a relatively clear definition of the researched “desirable” level of traffic.

Toll and Congestion Pricing

To clarify the conversation, it is critical to characterize what discriminate congestion pricing of other road tolling types. the report is at best fuzzy, and seems to mention the Port Mann and Golden Ears bridges toll as examples of congestion pricing. They are not, they are here to finance an infrastructure. A typical Congestion toll varies according to the level of congestion -and eventually become free on low level of traffic (night and week-end). below are the 3 main types of road pricing.

tolling type Main objective toll structure classification typology
financing finance an infrastrcuture vehicle size | weight classification
environemental reduce pollution vehile emission classification
congestion reduce congestion time of day/day of week

It is also good to feature the different congestion pricing coverage types:

tolling type tolling method example
infrastructure vehicle using the infrastructure Sydney harbour bridge/tunnel, HOT
cordon vehicle crossing the cordon Stockolm, Milan
zone vehicle moving inside a zone London

Some other more comprehensive or “ideal” coverage, mainly relying on GPS, to price the exact trip could be under study, however, so far we are unaware of a deployed one: let’s keep down on earth and rely on proven collection methods

Congestion pricing scheme around the globe

it is good to come with a general overview of what is existing today, to identify some possible trend, category, all this to identify more successful model, but also to get a better understanding of some failures (the failed experiment of Hong Kong, modeled after the successful one of Singapore should deserve a explanation):

urban toll overview, graph/classification from (1)

urban toll overview, graph/classification from (1)

the graph above purposely excludes most of the american city using HOT, or the Toronto ETR407 (at considering its purpose is to relieve congestion, what is open to discussion), under the rational they are usually located in a suburban context

Congestion toll Effectiveness

Cordon or zone tolling tend to significantly reduce the congestion by 15 to 85% [1], the case is more moot for infrastructure tolling.

In short: The Toronto ETR407 could have lot of merit and purpose, but general experience shown such type of infrastructure doesn’t reduce existing congestion

A recurrent and unfounded critic is that congestion toll revenue doesn’t cover the toll collection operation. In that instance the Eco-fiscal report would have misreported some numbers on the Copenhagen congestion pricing operation (see [5] for more detail on it). However the trend is toward lower operating cost.

Case studies

In the context of Vancouver, Stockholm is an interesting case study since it seems to share many characteristics with Vancouver, and its congestion pricing model can be transposed to Vancouver. However, it could have been interesting to study the difference, such as in term of transit ridership, urban form, and tolling area.

toll area surface and ratio tolled area/urban area

toll area surface and ratio tolled area/urban area

For reference Vancouver alone is already 115km2: a cordon pricing using the Fraser crossing could make the Vancouver area the greater area subject to congestion pricing.
The implication of this could deserve further investigation. Notice that, in theory the bridges surrounding the downtown peninsula doesn’t support enough traffic to justify a congestion toll [6]


Beyond the technicalities of road pricing, the acceptability need to be studied: Stockholm having passed the test of the referendum after experimentation is a good starting point: [7] gives some pointer toward acceptability.

  • Benefits may turn out to be larger than anticipated. Several authors have noted that a major reason for the resistance to congestion charges is that they assume they will not work.
  • The downsides of charges – increased travel costs and/or changes in travel behaviour – may prove to be not as bad as expected. Once the charges are in place, many people may discover that the charges do not in fact affect them as much as they had thought
  • Once the charges are decided, resistance may decrease due to the psychological effect known as cognitive dissonance. A phenomenon that can be simply summarized as “accept the unavoidable”. In other words, once the charges are in place, it is less worthwhile spending energy on opposing them.
  • Familiarity with road user charging may reduce the general reluctance towards pricing a previously unpriced good. There is evidence that “people in many cases do not like prices as an allocation mechanism”, but once familiar with the concept that road space is in principle a scarce good that can be priced – much like parking space – this reluctance may tend to decrease.

One will notice that the above paradigm would apply to bike lane too!

[7] mentions also the “environmental card”. It should also be noticed that the transit offer has been increased in the experimental phase, however most of the added transit is “self financed” by increased transit revenue from higher ridership according to [5].

Milan and its referendum

The “environmental card” should bring us to Milan, Italy, where road pricing has originally been introduced in the objective to reduce the pollution. However in 2011, Milan got a referendum. the question was

Would you like to extend the toll zone to the whole city and to all vehicles categories to fund transit and other sustainable alternative to the car?

The answer has been a resounding yes: 80%!

Needless to say, Vancouver has one thing or 2 to learn from Milan, …and we could have expected some element of response from the Ecofiscal report.


The Ecofiscal report is short on the technicality and specific of the proposed congestion pricing sheme. It could have been good to past the generality and get some hard and substantiated numbers to help the conversation.
While the Ecofiscal report apropos suggests experimentation, it doesn’t substantiate the rational for it (more especially its importance for public acceptability). More generally, the report falls short on recommendations enabling the acceptability of road pricing by the public. It doesn’t bring a new narrative making the whole congestion pricing concept more “sellable” to the “commons”. However, its main merit has been its ability to gather tremendous media coverage triggering conversation on road pricing: so it is probably good enough

[1] “Etat de l’art sur le péages urbains” (in french), Ademe, June 20142014

[2] Flikr user “Howard Brier

[3] “The Pedestrianization of Historic Istanbul

[4] Road Congestion Pricing In Europe: Implications for the United States H. W. Richardson and C. H. C. Bae , Edward Elgar Publishing, 2008.

[5] see the Cost-benefit analysis of the Stockholm congestion charging system, Jonas Eliasson, Transek AB, 2009. for more detail. This is inline with the used reference [4] in our 2011 post on the toll economics.

[6] see our previous post

[7] The Stockholm congestion charges – four years on. Effects, acceptability and lessons learnt, Maria Börjesson and al.

Good changes at Translink

October 14, 2015

In a transit plebiscite post-morten post, we were asserting that Translink needs to be much more aggressive in the optimization of its transit network, by going beyond shuffling bus around, and more noticeabily we suggested that:

  • A generalized bus stop consolidation policy needs to be initiated
  • costly detour, like the Champlain Height diversion on the route 49, need to be discontinued…

We are hence delighted by the scope and magnitude of the proposed changes by Translink. The thrust of it is converging toward a geometrically more efficient transit network, allowing to build a sound foundation upon which expand service. Here a quick review of some changes.

Moving toward the B-line paradigm
by retaining the main important feature of it; limited stop bus route:

  • route 135 is already a B-line, if not in name
  • high frequency leads to bus bunching on route 41: redeploying some of those buses in favor of the express route 43, will enable to offer more along 41st at no cost, while not necessarily decreasing the “useful frequency” of the current route 41
  • the idea to keep long haul buses (White Rock – King George station) serving all stop while short haul buses (96 Newton-King George) have limited bus stop, was an error. The introduction of limited bus stop along the 394 allows to correct this error: one will also notice this route goes directly to King George (the introduction of a LRT between King George and Newton could then be detrimental to White Rock)
  • the discontinuation of route 258, allows the consolidation of all downtown-UBC trip on a more legible route 44 (a clear candidate for a B-line title)

Other interesting changes of note are the revamping of bus routes around strong anchors.

bus routes tied directly to strong Anchors

  • bus 257 going directly to the Horshoe bay ferry terminal, instead of wandering in the Horseshoe bay village: Translink seems to have well learned of the private and successfull Bowen express bus initiative: the ferry terminal is the real reason d’etre of those services – It is up to the concerned municipality to capitalize on it by, for example, developing a cycling infrastructure to make those express routes more accessible to the local population.
  • The rerouting of the bus 509 from Surrey Central to Lougheed, is also providing people from Walnut Grove more accessible regional destination, more noticeabily, by providing acess to both Expo && Millenium line (including the Evergreen line extension) in a single transfer. It also contributes to relieve the King George branch of the Expo line, enabling a better use of the existing skytrain infrastructure.
  • the removal of the costly Champlain heigh diversion on the route 49 participates of this focus on building a strong network, with direct bus route linking strong anchors. We are glad to see Translink not giving up in face of the strong hostility of the Vancouver council to efficent transit.

The downtown case

The promotor of the new Vancouver Art Gallery, preserve transit service on Canbie street

The promotor of the new Vancouver Art Gallery, preserve transit service on Canbie street

The fate of the bus on Robson square is still open:

As outcome of the downtown bus service review initiated in 2013, TransLink does recommend the City to consider maintaining transit service through any future public plazas: that includes Robson and Cambie. Something we have long advocated for to preserve a legible and efficient transit network. The Translink conclusions don’t surprise us, and the proposed new route for bus 5 and 6 fit roughly in our analysis. The new layout of buses along Hasting/Powell seems also a step in the right direction.

Porch parade, was separating people on Robson square by a wall in 2015... Something a bus passing every 7mn or so, couldn't achieve.

Porch parade, was separating people on Robson square by a wall in 2015…
Something a bus passing every 7mn or so, couldn’t achieve.

As many, we have also noticed the failure of Viva to activate Robson square in despite of tremendous efforts. The 2015 choice, Porch parade, erecting a wall separating people rather than bringing them together, was not necessarily an happy one, but we have noticed a steady and inexorable decline in the attractivity of Robson square pre-dating the Porch parade experience. It is not surprising that the downtown BIA is now calling to restore good Transit access along Robson [1], which also would prefer to see the city move its attention on the North side plaza of the VAG. We are hence hoping reason will prevail at the Vancouver city council: It will then endorse the Translink recommendations.

And Change beyond Translink
In our post morten blog, we have underlined that the transit response to the lost Plebisicite must be two prones:

  • Rationalization of the Transit operations and network is one of them. Translink responded present (and in this blog, there is still  many other suggestions , such as the prunning of route 3 and 8…showing there is still room for improvement).
  • measures able to improve Transit efficiency, speed and reliability is another response under control of the municipalities

We commend the efforts of slow street toward the introduction of permanent bus lanes on Georgia street, but so far we have heard little more than politician rhetoric of our municipal leaders.

[1] Robson Street pedestrian space loses admirers, Kevin Griffin, VancouverSun, Agust 27th, 2015

The Hong Kong LRT

January 8, 2015

The topic of this post is the Hong Kong New Territories LRT, not to be confused with the double decker Hong Kong tram

The french approach: Some remarks
We have examined the french LRT approach and its history in a 2012 post: Some call it the art of the insertion to design Complete Streets [1][2].

It is also important to note the Certu (a French agency) limit for optimal surface operation conditions (signal preemption possible, perfect interval maintained) [4]:

  • A minimum frequency interval of 3mn. (in practice very few french LRT operate as les than 4mn interval)
  • A maximum capacity of 6,000 persons per hour per direction (pphpd)

The 2006 Paris Tram T3 ridership prediction didn’t forecast more than 3,800pphpd, and an average trip length of 3.5km (click for larger picture- source [3]

Indeed most french trams offer a capacity lower than 6,000ppphd and frequency interval greater than 3mn.
The parisian tram T3, probably the french busiest Tram, while having an excess of ~250,000 boardings/day, is not carrying much more than 4,000 pphpd [3] (that is almost the today transit capacity on the Vancouver Broadway corridor). Another important metric is the average trip length, it is 3.5km on the Parisian trams. For comparison the average transit trip length is 8km in Vancouver. The shorter the trip length is, the less important is the travel speed.

The Hong Kong LRT

The Hong Kong LRT illustrates it is certainly possible to go beyond the french limits, but it also illustrates the issues that entails.

The MTR suggests than in theory their LRT can carry 33,000pphpd, to recognize in practice that is not achievable due to street crossings. It is also worth to note than the MTR uses a very optimistic standard of 6person per sq meter: a crowding level Hong Kongers accept less and less. That said, it is probable that the Hong Kong LRT carries an excess of 10,000 pphpd on some corridors:

traffic and pedestrian congestion, slow (<15km/h) LRT prone to bunching even in their own ROW, all this in the middle of a high rises environment

traffic and pedestrian congestion, slow (<15km/h) LRT prone to bunching even in their own ROW, all this in the middle of a high rises environment

A first myth the Hong Kong LRT helps to dispell, is the supposed relationship between a transit technology and a building form. A myth unfortunately still alive in too many urbanism circles.

It also allows us to verify that a LRT can’t achieve any significant speed, with high frequency

  • The LRT is operated by line of sight: a frequency close to 1mn is possible, and is essentially limited by station dwelling time.
  • We have measured the average speed on the route 761P, which run in its own ROW, at less than 15km/h (this off-peak period)
    Due to the high frequency, LRT vehicles bunching are frequent and signal preemption not possible. Narrow crowded platform make also for long dwelling. For reference, the 99B average speed is around 20km/h

But the main issue is one of crowd management:

crowd management at level crossing, strict jay walking regulation, and ultimately grade separated platforms access, are all consequence of high LRT patronage

LRT and bike in Hong Kong
Where the street ROW allows, bike track exists along the LRT:

bike path along the LRT

A bike path along the LRT, notice how the pedestrians are prioritized on bike using a narrowing at the interaction. Also worthwhile to note that the LRT average speed is no more than 15km/h in spite of having a protected ROW on its full route length

and cycling is a well used mode to access the LRT stations:

bike parking at a LRT station

bike parking at a LRT station

However, while pedestrians get priority over cyclists as they should, the interaction between the both mode can be a bit clumsy, in fairly high traffic area:

narrowing bike lane, give pedestrian priority over bike at crossing

narrow sas, give pedestrian priority over bike at pedestrian crossing, but complicate cycling

Was the LRT the right choice?

As we have seen before, the quality of the urban environment is of little concerns to the Hong Kon authorities: The already anachronic high floor design of the LRT, at time of its opening in 1988, reflects this lack of concern. Poor LRT platform design and potential overuse of overpass is certainly a HongKong trademark too. Nevertheless, they have a “geometry” issue to address, which is the consequence of potentially too high pphpd for a nice integration in the urban fabric.

The Hong Kong LRT (top left on the map) is a feeder to the MTR West rail line

The Hong Kong LRT (top left) is a feeder to the MRT line

Since the opening of the West Line rail in 2003, the LRT is not a backbone transportation mode anymore, but a feeder to the mass transit line accomodating longer travel. That makes the LRT a right choice in regard of the short trip pattern to accomodate

That said, Sydney is building a LRT supposed to accomodate 9,000pphpd on its downtown segment, but

  • this segment (George street) would be fully pedestrianized
  • A single LRT line allows to operate long but less frequent trains, at least resolving some potential operational issue and trafic interaction

[1] Complete Streets: From Policy to implementation G. Thomson and T Larwin and T. Parkinson, TRB subcommitte on International Light rail develovpment – Rail-volution, Mineapolis, Sept 22, 2014.

[2] got it thru rail for the valley blog and Stephen Rees. We address here a question in a MB’s comment on the Stephen rees blog regarding the frequency of the French LRT

[3] Dossier du débat public: extension du tramway (T3) a Paris, – janvier 2006, Paris

[4] Tramway et Bus à Haut Niveau de Service (BHNS) en France : domaines de pertinence en zone urbaine from Transport/Environnement/Circulation (TEC) n° 203, September 209.

To better understand what bring the Mayors council plan (called “expansion plan” below), we ignore the spin and prefer to compare it with the Translink 2014 base plan (what is ensured to happen disregarding of the “plebisicite” result)

Congestion and gas tax

Fiat striking point: both plans estimate exactly same revenue for both the gas tax and parking tax. That is an implicit recognition that the Expansion plan will have no traffic impact, and per extension congestion impact (or if it does, it is mainly by the introduction of the Pattullo bridge toll): something we have already mentioned before.

Capital investment: $7 Billion above the $3Billion already included in the base plan

The $10 billion Capital funding is expected to be financed as below:


(*) The Pattulo bridge revenue is estimated from the 2024 operating budget ($50M/year) [3].
Notice that the figure doesn’t include debt service:

The “Congestion Improvement Tax” (CIT) finances ~22% of the capital funds needed.

Funded Operation (including debt servicing)

Revenue stream


Base numbers (e.g. “Transit revenue”) are presented for the the base plan, and increment numbers (e.g. “Inc. Transit revenue”) represent the additional revenu provided by the Expansion plan. the bump in 2017 is due to the sale of the Oakridge transit Center (planned in the base plan…but forgotten in the Expansion plan)

The original Expansion plan was targeting to raise $2 Billions over the next 10 years from a new tax to be triggered in several stage. The mayor having elected a 0.5% PST, will allow to raise ~2.7 Billions [1] over the next 10 years, creating lot of room for a more aggressive implementation that originally envisioned.

That said, at the end of the 10 years period, it looks like the PST revenue align with the original plan forecast.

Transit operation: $1.5B added on 10 years

In the next 10 years, the plan is apparently to put 400 more buses on the road, that is increasing the bus fleet size (actually ~1400) by ~30%…to increase service by 25% -it could be an issue here we will certainly revisit.

This, and other rail expansion services, will translate into an additional $1.5B of operating cost (including Transit police and Translink corporate overhead), generating $237M of additional transit revenue [2] as computed on 10 years : The new CIT tax, and additional senior government contribution (UPass) is expected to cover the $1.3B shortfall

Expanded Transit operation represents a relatively marginal increment on the base plan, but mainly funded by tax

Expanded Transit operation represents a relatively marginal increment on the base plan, but mainly funded by tax

The farebox recovery ratio of the added service is anemic [2]:

fare box recovery is expected to go up to 62% in the base plan. it will be 53% in the Expansion plan, thanks to an anemic 17% farebox recovery on the added transit services

Operation vs Capital Investment
In the first 10 years, nearly 50% of the expected CIT revenue will be devoted to operation (it could have been much more in the original plan). The partition look like below

nearly half of the CIT will be dedicated to operate the added service

In 2024, more than 70% of the CIT will be devoted to operate the added transit services, which will have a disastrous 17% fare-box recovery in 2024. That could even compromise the ability of Translink to pay back its debt, according too the CIT variation (inherently very sensitive to the economic climate).

It is possible that, some expanded service could pick-up steam in the years following 2024. If not, it looks those expanded transit are not sustainable in the long term, and will keep Translink on a train wreck course

That said, it is possible that our assumption on the PST growth rate is too conservative (the growth rate of the Metro Vancouver PST tax base is probably greater than 4%, but we have no solid number at this time)

[1] we assume a growth rate of 4% for the PST revenue. That is a conservative estimate, the PST growth rate province wide has been ~5% since 2008.

[2] we haven’t included the provincial contribution to the Students pass program

[3] the $1 billion figure represents the amount of debt which can be reasonably reliably financed by the Pattullo bridge toll. The Pattullo toll revenue forecasts are much more reliable than in the Golden Ears bridge, since it is an infrastructure upgrade

The Metro Vancouver mayors council plan, proposed to a 2015 referendum, calls for $765 millions of expenditure on the Expo an Millennium line over the next 10 years. This could result in an increase of 50% of the vehicle fleet and skytrain operating cost: Are those investments justified or just an extravaganza?

As of today, the Skytrain comfortably copes with the demand, thanks to the recently added vehicles in the years leading to the 2010 Olympic games, and should be able to serve the Evergreen line without hiccups, considering the expected addition of 28 cars. In fact the vehicle productivity (measured as rider/vehicle) is 20% lower from its 2008 peak. When the average increase in vehicle capacity is considered (83 passengers, before 1999, to 108 passengers in 2014), Skytrain vehicles productivity is at a 20+ years low (see our spreadsheet for detail).

To define the fleet requirement, Let’s see what the future ridership is planned to be:

Ridership prediction [1]
without a Broadway subway

2041 AM peak hour transit flow (without a Broadway subway)

2041 AM peak hour transit flow (without a Broadway subway)

..and with a Broadway subway

2041 peak hour transit flow with a Broadway subway up to Arbutus

2041 peak hour transit flow with a Broadway subway up to Arbutus

[1] doesn’t give explicit peak hour numbers for year 2021, but we can still infer them from [1] and [5] for the year 2021:

Maximum passenger per hour per direction (pphd).

without Broadway extension 2021 2041
Millenium Line 8400 10000
Expo Line 16000 23100
with a Broadway extension 2021 2041
Millenium Line 10400 12600
Expo Line 16000 19000

Thought the above projections could not have factored other transit investments such as the Surrey LRT or B lines, as contained in the Mayors council plan [3], they are not expected to significantly affect the peak pphpd requirement on either the Expo or Millennium lines.

The actual skytrain fleet is composed of

  • 150 MK1 cars.
    The 114 oldest car are currently refurbished, for an estimated amount of $38million [2], providing them an additional 15 years life span, so they are good to go up to ~2027
  • 108 MKII cars + 28 cars to be delivered in 2016 (Evergreen line).

The below table illustrates the usually used consists and associated train capacity:

4 car MKI train 4MK1cars-consist
332 passengers/train
6 car MKI train 6MK1cars-consist
498 passengers/train
2 car MKII train 2MKIIcars-consist
256 or 264 passengers/train
4 car MKII train 4MKIIcars-consist
512 or 528 passengers/train

We place ourselves in a scenario post Evergreen line:

  • The Expo line operates from WaterFront to King George (one branch),and to Lougheed (other branch): that is also called split-tail service by [2]
  • The Millennium line operates from VCC to Douglas college
The 3 skytrain lines, after integration of the Evergreen line spur

The 3 skytrain lines, after integration of the Evergreen line spur

Thought we are aware that Translink is considering to extend the Expo branch from Lougheed to Production Way, we are not considering it for the below reasons:

  • It doesn’t make good use of the skytrain capacity due to the poor expected ridership on the considered section
  • It creates operational and reliability challenge, due to the meddling of the Expo and Millennium operation
  • It significantly limit the capacity of the Millennium line: this one could be not required in the short-term, but discontinuing a service people get use to consider as granted, could prove to be troublesome in the future

2021 Rolling stock requirement

  • As per [2], we assume a minimum 93s headway and a 87mn round trip on the expo line and 78mn return trip on the Millenium line. Due to the ill designed Lougheed station, headway below 108s on the Millennium line could be challenging.
  • The extension of the Millenium line up to Arbutus increases its round trip by 15mn [1], and increases the pphpd requirement to meet by 2021, from 8000 to 10400.
  • We don’t consider short trains such as Commercial (or Metrotown)-WaterFront. They could still be used to reduce the fleet requirement or increase the spare ratio. Such strategy is not without issues [6].

No ext Broad. ext
Expo line Desirable (2021)
headway 93s 114s 114s 120s
train requirement 56

(31 4xMKII cars
25 6xMKIcars)
(21 4xMKII cars
25 6xMKIcars)

9 5 cars MKIII consists
12 4xMKII cars
25 6xMKIcars)

15 4 cars MKIII consists
26 4xMKII cars
1 6xMKIcars)
capacity (pphpd) 19,900 16000 16000 16000
Millennium line Desirable (2021)
headway 150s 120s 108s
train requirement 32
(32 2xMKII cars)
(40 2xMKII cars)
(36 4xMKI cars
16 2xMKII cars
capacity (pphpd) 3,000 6,000 8,000 10,600
Total Desirable (2021)
train requirement 150 MKI cars
136MKII cars
150 MKI cars
136 MKII cars
150 MKI cars
136 MKII cars
9 5 cars MKIII consists
150 MKI cars
136 MKII cars
15 4 cars MKIII consists
~10% spare ratio 6 5xcars MIII 8 4xcars MIII

The Broadway subway extension will involve at least the command of 7 new train consists (6 train consist to operate the segment + one spare)[1] which will be accounted as part of this project. So the extra rolling stock required to continue to meet the demand on the Expo and Millennium line in the next 10 years is:

Without Broadway ext. With Broadway extension
15×5 car MKIII consists 16×4 car MKIII consists
$262.5 millions $224 millions

the refurbishing of the remaining 36 MKI cars, estimated at $10 millions from [2] need to be added.

In the case of the Broadway extension, all other Expo line upgrades are already financed (federal gas tax subsidiary) and continue to carry on on schedule, so that the non yet financed cost is ~$240 millions (some minor egress improvement could be required here and there, especially on the Millennium line))

Potential additional storage requirement should be seen in the context of the Broadway extension project: The Coquitlam vehicle storage facility should apriori be expanded to accommodate, with the Burnaby OMC, the fleet up to 2031 [7].

Regarding the 5 and 4 cars consists

  • If the Broadway extension is not built, the expo line will require 5 cars train consist before 2041, so it eventually makes sense to consider to start to add such trains on the rolling stock from now, but that supposes also ancillary cost to adapt the line and the OMC, to longer trains it could also require upgrade of Waterfront and Stadium station, which are not yet funded. It requires also an upgrade (stage 3) of the propulsion power to enable the delivering of 25,000pphpd [6].
  • If the Broadway extension is built, there is no need for 5 cars train in the next ~30 years or the usual lifespan of a train: 4 car trains (MKII and MKIII generation) will be able to absorb the 2041 demand, and the line is already prepped out (or upgrade funded).

In any case, what should be ordered are trains able to maximize the capacity at a given length: The idea to order 3 cars train is a flawed one, since it doesn’t allow to realize the maximum train capacity, but more importantly prevent platform door installation (due to train assymetry making train doors location not always the same):

It is more than time to order rolling stock which will:

  • enable future platform screen, since such installation allow much greater system reliability than the current passive track intrusion detection model.
  • minimize dwelling time

That should imposes constraint on the train door location for any future procurement.

4 car MKII train 4MKIIcars-consist
512 or 528 passengers/train
4 car MKIII train 4carsMKIIconsist~540 passengers/train
3+2 car MKIII train 3and2MKIIcars-consist
~670 passengers/train
5 car MKIII train 5MKIIcars-consist
~680 passengers/train

~2030 Rolling stock requirement

Circa 2030, the original 114 MKI car will reach their end of life, as well as the 60 MKII (ordered for the opening of the Millennium line). we place ourselves in a scenario where those cars are still in service, and before a decision is done regarding their eventual life extension or replacement

By that time, the Expo line should be able to carry ~18,000pphpd and the Millennium line, ~12,000pphpd (number inferred of both the 2021 and 2041 projection). The rolling stock could be assigned as below:

Expo line (2030) Millenium line (2030)
headway 108 150
train requirement 48
26 4xcars MKIII consist + 22 4xMKII cars
25 6xMKI cars + 12 4xMKII cars
capacity (pphpd) 18,000 12,000

Considering a ~10% spare ratio, 36 new 4 cars train should be ordered by 2030. More likely 30 in the next 10 years with an option to order 6 more circa 2025. That includes the 7 train part of the Broadway extension project, so the effective requirement could be 29 4 cars train – or 23 train in the next 10 years period, that is ~$320M (with a 6 additional 4 cars-train option to exercise ~2025)


  • the possible availability of second hand MKI car (from the Scarborough RT or the Detroit People Mover), and potential acquisition for refurbishing should be considered
  • the decision to go with 4 or 5 car consist order should be reexamined in the next 10 years, in light of the ridership evolution

The Mayor council plan

In brief the Mayor council plan[4] calls for the below

% increase
increased operation cost 53.5 50%
capital cost $765 millions
new vehicles ($500 millions) 145 50%

The above doesn’t account for 27 vehicle to be procured between 2025 an 2029

In the light of the previous sections, this seems to be an inconsiderate expense to

  • address purposeless goals; such as doubling the capacity of the Expo line by 2020 (the main reason for the mayors plan extravaganza)
  • and still failing to address basic requirement, such as the 10,000 pphpd ridership on the Millennium line in the case of the Broadway line (the Mayors council’s plan consider only 8,000).

The Mayors council’s plan implicitly assumes 3 cars train: This is a bad idea as we have seen before

A fundamental reason to put the Broadway subway as the top priority transit investment is to spare the considerable expense to upgrade the Expo line to meet the ~23,000pphpd 2041 demand; which could happen only on the very short section Commercial-Stadium:

A Broadway subway will reduce the Expo line demand at ~19,000pphpd: something achievable as of today, and could save ~$300 million of investment on the Expo line, according to the council mayors numbers [8], and associated operating cost, otherwise necessary.

The fact that the passenger load is much more balanced along the Expo line, in the case of a Broadway extension, make a much better use of the line capacity.It is still possible to operate short train in the other case, between Commercial (or Metrotown) and Waterfront, but it doesn’t come without issues ([6]), such as passenger bunching or platform crowding (due to passenger waiting for the expected less crowded short train)

It is unfortunate the Council of Mayors missed this important point.

[1] UBC Line rapid transit study: Phase 2 Evaluation report Steer Davies Gleave, August 2012

[2] Translink 2013 Business Plan Operating and Capital Budget Summary

[3] Regional Transportation Investment: A vision for Vancouver – Appendices, Mayors council, June 12 2014

[4] Regional Transportation Investment: A vision for Vancouver – Appendices, Mayors council, June 12 2014

[5] TransLink’s Rapid & Regional Transit Model , PTV America Inc. and Translink, Vancouver and Wilmington, DE, February 2007 and December 2008

[6] Expo Line Upgrade Strategy, SNC Lavallin and Steer Davies Gleave, Sept 21, 2010

[7] We estimate the current storage capacity at 114 MKI + 126 MKII at the Edmonds OMC, 36 MK1 and 34 MKII on the main line an the Coquitlam Facilities storage center. See the Translink Finance Audit – Specific Project Approval. Subject: SkyTrain OMC Expansion – Phase 2. October 19, 2007 and an ensuing discusssion on the skytrainforsurrey blog

[8] That is the difference between the Mayors council plan, $765M and our ball pack numbers, $320 for rolling stock expansion/upgrade and ~$150M for infrastructure upgrade, including storage/OMC expansion: Those numbers are in fact consistent with [6]

Below we provide mainly some number coming from [1], using a data base; the “registre du Rhone” (Lyon district, France); recording detailed information for 14 478 cyclist victims of road accident in the Rhone county (France). The data under studies have been collected from 1996 to 2007. The level of detail is very high compared to some other data base, and they can help to clarify the helmet debate:

see wikipedia for the AIS and IIS definition and meaning

Injured Body part among cyclists
reported numbers are only for cyclist where the helmeted/unhelmeted status is known

helmeted 1461 unhelmeted 5620 helmeted 1461 unhelmeted 5620
Body region %yes %AIS3+ %yes %AIS3+ yes AIS3+ yes AIS3+
head 16,6% 1,0% 19,5% 2,3% 242 15 1095 128
face 18,3% 0,1% 24,3% 0,2% 268 1 1367 9
neck 3,3% 2,7% 0,0% 48 0 152 1
Thorax 9,6% 1,8% 6,3% 1,0% 140 26 355 55
abdomen 3,9% 0,1% 3,6% 0,2% 57 2 200 14
spine 7,5% 0,3% 5,0% 0,1% 110 5 281 8
Upper Extremity 53,5% 3,2% 44,7% 3,4% 781 47 2513 189
Lower Extremity 33,7% 1,8% 31,2% 1,9% 493 27 1751 107
skin 17,8% 15,2% 260 0 857 0

This table shows that the helmeted cyclist are less often severely injured to the head, less often injured and seevrely injured to the face. But they are more often injured and severely injured to the Thorax and spine, and more often injured to the upper extremity. This suggesthelemeted cyclist get involved in more severe accident. 2 reasons can be advanced for it

    • Helmeted cyclists have a different cyclist practice that non helmeted one (that relate to the infamous MAMIL)
    • There is a risk compensation at play along helmeted cyclist

the effect of it is to give an over estimate of the helmet efficiency

The figure below illustrates the % of head injuries oberved for injuried cyclists with an helmet and without:

The bike helmet seems to have no significant effect at reducing superficial head injuries but it reduces the risk of severe head injury (AIS3+) by 50%. However such severe head injuries represent less than 2% of the total cyclists injuries

So we are looking at a global effect of 1%, but severe head injuries can have long term disabilities consequence, so a reduction by 50% of severe head injuries could be considered a worth endeavour:

Below a more detailled view of the head and face injuries specificities:

Head or face Injuried cyclist: type of lesion according to helmet wearing or not

Helmeted 238 Unhelmeted 1078
lesions N = % N %
Unconsciousness without additional descripted lesion or head trauma without further indication 194 73,5% 668 49,3%
skin lesion 43 16,3% 459 33,9%
brain 21 8,0% 150 11,1%
skull 6 2,3% 65 4,8%
little brain 0 4 0,3%
nerves 0 4 0,3%
brainstem 0 2 0,1%
Destruction skull and brain 0 2 0,1%
Total 264 100% 1354 100%

In despite of a pretty huge database, significativity is not reached on many lesions types.

According to [1], Among the 222 cyclists injuried to the skull or its content

  • 30 are dead
  • 42 will have severe heavy brain disabilities (IIS3) – that represents 0.3% of the injuried cyclist
  • 146 will have light brain disabilities – that represents 1% of the injuired cyclist

The effect of the helmet on long term disabilities seems to be measured on a very small cohort:

unhelmeted 35 helmeted 3
fatalities 15 0
%fatalities among total injuried 0,3%
%fatalities among head injuries 1,4%
Severe head disabilities (IIS3+) 20 3
% Severe disabilities among total injuried 0,4% 0,2%
% Severe disabilities among head injuried 1,9% 1,3%
% Other severe disabilities in addition of severe head disabilities [10] 19,8% 21,4%

Again, we don’t reach significativity to be able to conclude that a “bike helmet saves life”, but it could well reduce significantly the risk of severe head disabilities. The number could suggests 50%, but again with very little confidence.

In anycase, the effect is measured in tenth of 1% of the total injuries: the absolute number of victim is very small.

How bike injuries compare to other transportation mode?

When involved in an accident, cyclists have the lowest fatailities rate compared to other individual transportation mode. They have also the best chance to survive an accident without long term disabilities. Urban cyclists are even much safer than rural or suburban cyclists:

Injury severities Uurban 6584 Rural 1726 0-10years 3364 Pedestrians 10131 car 53151 Mopette Motorbike 21831
MAIS 1 63,7% 53,7% 72,5% 60,4% 81,6% 61,7%
MAIS 2 28,7% 35,3% 22,4% 24,7% 13,4% 27,0%
MAIS 3 5,8% 7,9% 4,2% 10,0% 2,8% 8,6%
MAIS 4 and 5 0,9% 2,0% 0,3% 2,3% 1,0% 1,5%
fatalities 0,6% 0,8% 0,1% 2,6% 1,2% 1,1%
no disabilities 78,4% 73,8% 88,2% 66,1% 55,3% 73,7%
light diabilities (MIIS1-2) 20,2% 23,8% 11,1% 29,6% 42,8% 23,8%
severe disabilities (MIIS3-6) 0,6% 1,2% 0,2% 1,6% 0,7% 1,3%

When urban cycling is considered alone, helmeted cyclists are observed to have 20% less chance to be severly injured that non helmeted one, whenever involved in an accident, however significativity is not reached [1].

The figures below eventually help to compare the different pattern of injuries, according to transportation mode, and eventually relativize the impact of a bike helmet on the total number of both severe head injuries and severe disabilities.


Injuries type according to transportation mode (from [2])

Long term severe disabilities type according to transportation mode (from [2])- head injuries represents ~60% of the long term disabilities fro both pedestrians and cyclists

those figures provide absolute numbers for the Rhone district (1.7Millions) inhabitant for years 2007-2010…
Even if a bike helmet is able to reduce the severe head injuries and disabilities, by 50%, the saving in term of public health could be relatively insignificant but the more generalized economic cost can be real: It has became clear that an helmet legislation is counterproductive, but even promotion campaigns in favor of bike helmet could be counter productive:

  • The money spend on helemet couldn’t justify the health care saving, and could be better invested somewhere else
  • Helmet campaign promotions paint cycling as a dangerous transportation, discouraging people to cycle, conducting to a less healthy population

One could object that a low absolute bike injuries number is due to a relatively low cycle ridership. The dramatic increase in cycle ridership over the last decade, especially in urban area, is well documented, and still the number of total injuries has not increased but is decreasing [2]:

Bike accident evolution in France

Bike accident evolution in France

The more people bike, the more a bike helmet becomes irrelevant…unless you believe pedestrians should wear an helmet too!

More citation to come

[1] Vélo et casque, Journée spécialisée, jeudi 28 mai 2009, Rapport UMRESTTE 0908, May 2009

[2] Bilan 2011 de la sécurité routière en France

[10] doesn’t express it like it, but I am assuming it is due to a typo.


Get every new post delivered to your Inbox.

Join 61 other followers