Quote:
Originally Posted by sim
I'd be interested in the documentation for this if you happen to know where it might be / if it's public.
I don't think it's worth getting into details of classical 4 step modelling in this thread (I'm assuming the forecasting was done with this, as they usually are), but there are a lot of assumptions that can be made in them, to the point where you can get the answer you're looking for. Not to mention that they can be [are], largely self-fulfilling, their aggregate nature can be rather erroneous and their forecasting is based on past conditions that are assumed will hold, etc. etc.
I'd also speculate that essentially any route assignment method used is probably quite divorced from what will happen in reality. Going way around the entire SW portion of the ring road, it might be objectively faster in the model, but that amount of distance will seem subjectively longer to a lot of people (I'm assuming a NW to SE industrial area pattern).
In short, I don't buy the modelling if it somehow predicts that we'd need about 3 Deerfoots over there to meet "demand". Not until I see really extensive documentation and all assumptions.
For an interesting collection of models missing the mark (significantly), see almost every recent toll road project - where the modelling is supposedly rigorous considering they are somewhat used for the underlying financing and operating structure. We need only go as far as the Port Mann bridge.
|
It's normally the responsibility of the owner to follow up with volume recording post-construction. Considering consultants use data from local and provincial government forecasting to generate their simulations, it should come as no surprise that they get forecast data from the City of Calgary and Alberta Transportation to generate their models.
As far as I know, this is the extent of the public volume data available:
http://www.calgary.ca/Transportation...g-Toolbox.aspx
http://transportation.alberta.ca/3459.htm
In this very thread we discussed the early numbers that were coming out on NE and NW Stoney (I believe it was 6-8 months post-NE opening) and the O-D pair results. Didn't search this thread for it now but it should be around.
I believe SW Stoney was done with microsimulation tools, and to refer to the analogy made earlier, there's no way they designed it like a restaurant where the brunch never had a lineup. They don't design roads with the intent of it never slowing down in a peak period. Lower speeds and congestion are acceptable, what you're trying to avoid is the road failing. They don't alter the volumes so they get the results they want, if that's what you were implying. They change queue lengths at intersections, lane counts on ramps, change interchanges from Parclo A to B, things like that.
I have no problem with not relying on LOS in urban situations where the ability to expand is minimal, but that doesn't apply as well to SW Stoney, a greenfield development in a corridor devoted to utilities and high speed roadways.