Traffic Flows -What The Broadway Subway Boys & Girls Do No Want Us To Know
Mr. Havecow is a transit professional from Eastern Canada, who has studied our transit system and of course SkyTrain. His findings certainly point to the fact that the public have been inundated with porkies, big and small, from TransLink , The City of Vancouver, and other retread politicians and academics who are clinging to the SkyTrain subway dream. Until there is both honesty and clarity by TransLink and the city of Vancouver, we must assume that claims that would support a SkyTrain subway are dubious in the extreme.
“Remember, these big numbers for passenger flows that are being in my opinion, carelessly bandied about by people who want a subway are boardings not trips.”
Over to you Mr. Haveacow.
Many thanks to those whom answered my questions. Using the 2012 ridership for the bus routes that were provided to be by Translink then converting their boardings numbers into actually trips, as well as using the basic Logit Toronto B model for the calculation of flows and the CUTA Model for the calculation of various bits of info used in figuring out transit capacities, various job distributions and weightings, I came up with the following numbers about the Broadway Corridor.
Keeping in mind that, numbers from any mathematical model must be verified with on sight observations and the formula ranges adjusted and recalculated over a period of years to be considered fully true. Also that, I donai??i??t live in Vancouver so, observations and recalculations to match observed views has to be done by Translink, they have the staff and time, plus no one was paying me to do this for them so the week it took to crunch these numbers is my time lost (however anyone wishing to throw a donation my way should know that it will be gladly accepted by me and the bank that holds my mortgage).
The Broadway Corridor averages 56652-64153 trips per week day. The maximum daily trips number given with all of the variable inputs favoring transit was 75710 trips per day. Using the mean trips per day range, 27.8% of all trips in the corridor come from passengers transferring from surface routes. 65.2% of all trips are generated from transfers from either the Canada or Expo Rapid Transit Lines and roughly 7% comes from within the corridor.
Most likely because of the University, the PM peak period has a 2-10% greater travel flow than the AM flow. This is usually reversed for most North American cities. Remember that this is just for this corridor not the whole Vancouver Area so the normal transit trip distribution could still be true.
The distribution of trip origins and destinations is very muddled in this corridor due to the distribution of large trip generators like City Hall, Rapid Transit Lines, VCC, UBC, Vancouver General Hospital Campus (including the cancer centre), the density and distribution of commercial properties and the residential properties. Again I would like to thank Doug from Translink for his data sets in this regard.
Overall as was observed by Rico, a range of 28-42% of trip origins and destinations were west of Arbutus with a median value around 34.3% mainly due to the university campus. The information provided by Translink also showed that, the seat/standing room turnover rate peaks east of Arbutus, then sharply declines the closer you get to UBC. This means that the majority of pass ups by buses should peak the closer you get to UBC because fewer people are giving up their seat or standing room on the buses.
The data from Translink also showed an interesting thing about the functional capacity of the bus routes. I have argued before that anyone can with enough info, calculate the maximum capacity of a transit vehicle. Whatai??i??s really needed is how close does the service get to its capacity before the majority of passengers refuse to get on, look for other transport alternatives or delay their trip in some way to avoid peak travel times. I have been working on a model with primarily one other person that, given some standard imputs will predict this activity. What is shocking to me that income had very little effect on this as well as age. We seem to like are consistent travel patterns and are very reluctant to change them once we find something we are comfortable with. What is interesting about this corridor is that, on average once transit vehicle occupancy gets to about 75-83% of stated capacity, people traveling in this corridor start to find other choices. For an area like Vancouver that relative to other cities has an average range of 82-88% of capacity tolerance, which is high, this is very low level of crowding. It maybe due to the large amount of bags and packs that university students tend to have to carry on a daily basis but, that is a only a guess.
The amount of service in the peak periods varies in this corridor from 15 to 22 transit vehicles per hour per direction due mainly to the Trolley Bus network running buses on part of the corridor then going in other directions at various points. Due to that last fact the maximum passengers per hour per direction flow has a wide variance. As mentioned before the flows are slightly higher in the PM peak period (which is unusual) but still close. When transfers from other surface routes are considered the median range across the corridor varies from 1935-3705 trips per direction per hour. The maximum peaks were 4539 trips/hour/direction in the PM peak from Arbutus to Granville and 4705 between Main and Fraser during the PM peak hour. At no point did the flow exceed 5000 trips/hour/directions, on a standard work day. I thank all the people who helped me in this endeavor, there were many.
Oh yes, the flow diagrams showed that bunching of vehicles is quite common in this corridor which can give the impression of a greater number of vehicles in service than is actually true. Plus the nature of trolley buses being tied to their lines and having only a limited ability to by pass other vehicles contributes to the bunching. However, I have seen them move around smaller vehicles in heavy traffic in Vancouver as well as Edmonton and Toronto (when they had them). It all depends on operator confidence and pole length used.
Followed by this little nugget…….
One also must be careful about using future demand numbers as a reason for rail rapid transit. What my modelling showed was that the peak hour numbers are manageable by BRT or LRT lines and that a below grade light subway/metro is currently overkill. What needs to be addressed IMHO is the non peak crowding which can be handled by either more standard bus services or a conversion to a real BRT/LRT system on this corridor. Remember, these big numbers for passenger flows that are being in my opinion, carelessly bandied about by people who want a subway are boardings not trips. When you convert them into actual passenger trips the numbers become up to 40% smaller. I have seen people on this website and others claim that Broadway will garner up to 200,000 passenger trips a day by 2030, this is a ridiculous claim considering that, right now the entire system only gets 723,547 passenger trips a day (1,176,500 boardings a day). If Broadway currently accounts for less than 10% of the systems daily total then they are predicting a huge increase for the entire system by 2030. Now it is a busy corridor there is no denying it, but certainly not really ready for a very expensive below grade light subway/metro line. Even the Evergreen Line is only predicted to have 55194 boardings a day (17,000,000 a year) by 2021, thatai??i??s about 35,000 passenger trips a day. Considering the financial pressure both capital and operating expenses, that Translink is currently under, aAi?? 2-3 billion dollar line on a corridor that even when everything favors transit at most gets 75,000 passenger trips a day, but averages 56-64,000 a day is at best, not advisable at the moment. Perhaps a lower cost system like LRT or a real BRT line (not the current BRT lite system) is what is really needed in this corridor.
Oh Eric, I could not add this to the other conversation, I guess we hit the limit. Anyway here you go.
Sorry I guess it wasn’t clear enough the 4705 is passengers per hour per direction. However, it is a maximum point volume for a tiny part of the right of way which drops when the trolley bus route that runs along it veers on and off the corridor. Many of the high point volumes along the corridor are caused by another north-south bus route entering then leaving the corridor. Although there is already a relatively high rate of transfers coming from other surface buses to the routes running the length of the corridor. I was mildly surprised that, the percentage wasn’t higher due to the number of high volume trolley bus routes bisecting the corridor. It suggested to me that, many passengers take these other north-south bisecting routes because it travels the precise point on the corridor they need personally, so they don’t have to transfer to a cross Broadway route, when they leave their north-south bus. However, that level of analysis and associated statistical proof was way beyond what I really wanted and had time to do. Especially for 3 routes the 8,16 & 17. This is why I think a more precise modeling needs to be done to see where every single passenger is actually coming from. That way it maybe possible to do something to build more ridership and ease crowding on the corridor at the same time, without spending huge sums of money.
@Haveacow.
Thanks very much for the clarification; you did a much more thorough analysis than I realized. Now I get how TransLink is likely cooking up the exaggerated numbers for the sky train line up the Broadway corridor. These exchanges on the RFTV blog really do help my understanding and are invaluable for me to unravel the twisted thinking at TransLink.
From a design perspective for any rail line (tram, LRT or sky train) up Broadway, the north-south routes will not be replaced and their riders to a great extent do not count as far as the capacity required along Broadway. They are just unnecessarily clogging the Broadway corridor and the east-west travel for these routes can be shifted off Broadway to West 4th Avenue or West 16th Avenue.
We know from UBC data that only about 2100 pph reach UBC on the No. 9, No. 14 and No. 99 routes traveling along Broadway. Therefore, we can deduce that riders on the north-south routes aren’t transferring to the buses on Broadway to reach UBC in the morning, to any meaningful degree. Due to symmetry, we can say the same thing about riders as far as transferring to buses (east-west) along Broadway to reach Commercial Drive, also.
If you run a tram line or LRT line at grade on Broadway, you will have to change the east-west travel on Broadway for the north-south trolleybus routes (No. 3/8/16/17) to either West 4th Avenue or West 16th Avenue. In my mind the only real routes which will be replaced with the rail line up Broadway are going to be No. 9 and No. 99 (13.4 km along Broadway) and the No. 14 traveling to UBC along Broadway over 8.2 km. When you look at it in terms of the transit demand that is going to be replaced by the rail line up Broadway, it is about 2,500 pph along Broadway, it appears, from your modeling.
I’m reading in between the lines of your modeling. Perhaps, I’m not getting it right and the demand excluding the north-south routes is still over 2,500 pph. Let me know.
Yea it is but as you saw, not really that much over 2550 (using the average range anyway). Even the peaks really are not that high. I have a big feeling that, the crowding caused by the post rush hour service being lowered too much compared to the need coming in and out of UBC and the bunching of transit vehicles gives the false impression of higher usage to the rider. But, that would require more study, wouldn’t it ( and no one is paying me so its up to Translink or someone else).
To be fair to Rico, he is not unique in his confusion. The problem of modeling in isolation, a single corridor for example is that, you can use many methods to show future growth. Some are better than others but, it has the problem of examining a corridor in isolation. The problem of examination is always that the effect on the area the corridor is located in, none of the official modeling done by anyone really looks at the relationship of north-south routes that serve the corridor or the effective relationship between other areas or corridors.
If my maximum numbers for the number of passenger trips a day is correct, 75100 then by rough calculation the Broadway Corridor produces roughly 10% of the daily passenger trips for the whole system. If some of the future predictions of boardings by translink are correct than one of two things is beginning to occur. Either the fundamental relationship between this corridor and other areas of Vancouver including central Vancouver is starting to or is already massively changing. This will mean that downtown and all its high priced land is in serious decline relative to the Broadway Corridor. I just don’t think with all the development opportunities in greater Vancouver that, Broadway is that strong and overwhelming. You guys live there, is this an accurate assumption?
The second thing that may be occurring is that the huge, in some cases, 300% increase in boardings in this corridor is part of a massive system wide growth in boardings by 2046. I’m not saying that transit use will not grow but, some big fundamental changes to society must be being used in the modeling. Possible sure, but if growth does occur like this in greater Vancouver, I seriously doubt that, the Broadway corridor’s place in that growth is going to change so significantly compared to other areas. The models must also be predicting some big overall population growth as well.This is also very possible but, still a big change in the relationship between the Broadway corridor and the rest of the community is also predicted as well on top of the population growth seems to be implied.
The point is that when you go too far into the future when you model a specific corridor without really looking at the relationship the corridor has to the whole community than you may get very large numbers but, numbers that have very little to do with
Sorry accidentally hit submit button, let me finish, numbers that have very little to do with what can actually happen because of the disconnect in the modeling between the outside community and Broadway. There is most likely a name for this phenomena but it escapes me right now. Modelling a single corridor into the future is useless unless you can attach the many existing relationships that are already there inside the modelling.