It will be much faster than the current local buses:
Evidence suggests not... Round trips during AM peak for St. Clair after the ROW only went down by 8 minutes, so the maximum time savings riders saw was 4 minutes if they were riding the entire line. I was only quoting the widely available figures for average speeds, you're free to disagree with them.
-the current buses always get stuck in rush hour traffic, LRT is immune. Traffic will only get worse in this city.
It's actually pretty rare for buses to get "stuck" in traffic.
-when the bus is busy people get off every stop, it takes a long time to get people to pay their fare and go up the stairs, so dwell time per station is long. this is made worse by the fact that the buses are completely packed
All-door-boarding/POP and longer articulated buses (both of which are apparently forth coming) would mitigate this problem. None of this is inherent to LRVs. We've just chosen not to implement it on our existing LRVs and buses.
-the bus stops have MUCH closer stop spacing than the LRT, especially the central part of Eglinton, which multiplies the delay for each stop
Again, stop spacing is independent of technology choice. You could drive a TTC bus express to Vancouver, if you wanted. That said, the surface LRT ROWs are supposed to be roughly the same stop spacing as existing bus routes, no?
Having said that, the buses perform better at night at less busy hours, since there is less traffic and it doesn't stop at every stop. Also, you're right that an enhanced bus service with the same stop spacing as the LRT would be much closer to the LRT. More more you change the bus line to be like the LRT the faster and better it gets.
It's not changing buses to be "like" LRT. Things like boarding policies, stop spacing, and such are 100% separate of vehicle choice. We've chosen to run our existing surface routes (including our LRV fleet) in a horrendously inefficient way.
TheTigerMaster said:
It includes both peak and off-peak travel times. So that 18km/h average bus speed includes 11PM busses running at top speed with only a few passengers on board as well as busses that run at peak usage, when most riders use the system and the busses are stuck in traffic. So this methodology obviously inflates the numbers in favour of the busses, since the light rail lines will run at a reliably 22 km/h.
Yes, as I said, mixed traffic surface vehicles would probably have higher variance than vehicles running in reserved ROWs. That said, it's probably not as high as you think. Buses are pretty rarely 'stuck' in traffic. Even if you took a Porsche 911 on an empty race track and made it stop every 400-500m for 30-60 seconds, you're average speed would end up being only 50-60km/h. Try doing that with a bus or LRV hauling hundreds of standing passengers and it's easy to see why transit vehicles can't really exceed 25km/h on local routes, even if there was no traffic or signal dwell time.
There's lots of real world evidence for this. Look at Steve Munro's analysis of routes like the 54.
Peak period effects do exist, but they're usually contained to small parts of the route and modest. Or look at the evidence from St. Clair, where the ROW didn't lead to massive reductions in rush-hour vehicle speeds. Transit vehicles will tend to be slower than prevailing traffic, often even in peak times.
Further, if the issue is limited to peak hour, then the most efficient solution is most likely not new infrastructure but rather time sensitive restrictions. Judiciously enforced rush-hour bus lanes would likely have similar impacts to building LRT lines.
EDIT: Actually, a second follow on observation would be that express-local set ups could be more efficient than simply increasing station spacing a bit. For instance, add a service which would run non-stop from Finch Station to Seneca College then run local there on out. That oughta save transit riders East of the DVP ~10mins, and improve capacity utilization.