I have been catching up on some older discussions on this list, and this one really caught my eye.

The subject line suggests questions about dealing with traffic congestion on a dedicated busway that is functioning at capacity (as is currently the case in Ottawa, Canada and Brisbane, Australia). Although it is an interesting topic, it is not the nature of the questions that I have seen so far on this particular discussion thread. What I have seen are comments such as this one from Chris Cherry:

> There has been some research ... looking at trying to maximize 
> the use of these "empty" BRT lanes (or any limited access lane 
> for that matter).

Bus lanes on city streets or highways (as opposed to completely segregated busways) suffer from an image problem which is not really shared by rail tracks. In its simplest form, the motorist stuck in traffic in the adjacent lane sees the "empty" bus lane beside him and wants to use it in order to move forward faster. If there is a rail track (excepting tram tracks that are paved over to permit joint use by trams and other vehicles such as buses or cars), the motorist realizes that there is no possibility of using that space, and he is therefore resigned to his fate of being stuck in traffic.

The problem is that every user is concerned only about his own trip and not about the efficiency of the system as a whole. If you have 30 buses per hour using a bus lane with 50 passengers per bus, that bus lane is carrying 1500 people per hour, which is about the equivalent of a single lane of freeway or two lanes of non-freeway arterial road. Indeed, here in Ottawa, we have figures similar to that ("the bus lane is carrying as many people as the other two lanes of traffic combined on that road"). At 30 buses per hour, or a bus every 2 minutes, the lane is empty most of the time, which creates the pressure from the other motorists to open the lane up to more usage. Of course, the lane only serves its purpose if it is uncongested. As soon as it fills up with other cars, then it ceases to be a fast and reliable method for buses to move.

Until today (see below), my attitudes about shared use of transit lanes were as follows:

(1) If the only other use of the lane is for moving cars, then it should be a full-time transit lane. Signage and enforcement of part-time lanes is much more complicated than fulltime lanes. During off-peak hours (in the urban contexts in which I function), there generally are no traffic congestion problems, so the question is moot as to whether cars are allowed to use the transit lane or not. If there are congestion issues, then the buses (or other transit vehicles) need to have the lane available to avoid the congestion during those hours as well as during the peak hours. Either way, the result is that the lane should remain reserved for transit.

(2) If the lanes can be used for other functions, then there may be good possibilities for joint usage. The classic example of this is a lane that is reserved for buses during peak periods, and which provides parking for adjacent businesses during off-peak times. Generally this sharing of function can work quite well. 

(3) If the pressure to allow more traffic into a bus lane is too intensive, then make it an HOV (high occupancy vehicle) lane. The important thing is that the bus travel remain unimpeded. If too much other traffic starts blocking the lane, then just keep increasing the minimum number of passengers until the traffic moves freely again. If HOV-2 (minimum 2 people per vehicle) clogs the lane, increase it to HOV-3. If HOV-3 means too many vehicles, then increase it to HOV-5, or even higher. :-)

In the USA, some jurisdictions have converted their HOV lanes into HOT lanes ("high occupancy toll"). This means that solo drivers can use the lanes if they pay a toll, whereas the regular lanes are free (or a lower toll). As the ultimate "high tech" solution, the tolls are variable in real time so that the lane remains freeflow. If too many vehicles fill up the lane, the toll rises until freeflow is restored.

The problem I see with HOT lanes is enforcement. "Justice must not only be done, but it must also be seen to be done":

- The simplest lane to enforce is a bus lane. Both the police and the other motorists can easily see who is violating the lane. If you're not a bus, you get a ticket.

- Slightly harder to enforce is an HOV (carpool) lane. Now the police officer has to count the number of heads in a car before pulling it over. More important, perhaps, is the social stigma of violators in the lane. The average motorist sitting in the other lanes now sees vehicles comparable to his whizzing past him, and this will encourage him to violate the lane. It's not easy for him to know whether there are enough passengers in the other cars or not.

- The HOT lane adds an additional negative dimension to the enforcement issue. Not only does the police officer have a harder time determining whether the vehicle is legitimately allowed in the lane, but the motorist in the adjacent lane has no way to know whether the solo driver in the reserved lane has paid a toll or not. This encourages higher violation rates.

I would encourage transportation planners to be aware of these public perceptions, and to educate their politicians and public as best as they can about the issues. Let them know how many PEOPLE (not "vehicles") are being served by the transit lanes in comparison to the remaining traffic lanes. Point out that there may be dozens of parallel lanes available for cars along the corridor, not just on the individual street, so you haven't taken away half or a third of the traffic capacity just for buses, but one lane out of N lanes in the whole corridor. If it is necessary to open us transit lanes to other vehicles, then it should be to taxis and carpools only, and only to the extent that freeflow traffic is maintained.

Moving on to a different issue, as per these comments by Jonathan Richmond: 

> does anyone think that this principle could be 
> extended to optimizing the use of BRT rights of 
> way to allow buses to enter and leave at various 
> points without causing congestion?
> [...]
> I want to keep the lane to buses only 
> at all times, but to permit buses to enter and 
> leave at a variety of places without causing 
> congestion.

Jonathan, if I'm understanding you correctly, you're concerned about buses creating traffic congestion at the point where they re-enter mixed traffic. One thing to remember is: buses are never the cause of traffic congestion! They are part of the solution to traffic congestion! For every bus on that section of roadway, consider that the alternative would be 40 more cars (or 20 cars if they're lightly loaded buses in the USA, or 150 cars if you have crush-loaded artic buses). Which is easier: finding space for 1 more vehicle, or for 40 (or 20, or 150) more vehicles?

The access/egress points to/from a busway can be handled through normal traffic signals and other normal traffic engineering methods. The buses should be given preferential treatment at those traffic signals for reasons which should be obvious. One of the major problems in North America now is convincing traffic engineers to consider the total person-minutes of delay rather than the total vehicle-minutes of delay. If your road network is such that having a few buses enter it creates significant congestion, then perhaps that's an indication that you need to extend your busway even further until it is past the point of constriction.

Chris Cherry provided this reference:

> http://www.its.berkeley.edu/publications/UCB/2007/VWP/UCB-ITS-VWP-2007-1.pdf

and in the abstract I read:

"a freeway's [non-HOV lanes] discharge vehicles from bottlenecks at an equal or higher average rate when one of the lanes is devoted to high occupancy vehicles than when it is not"

This suggests that the starting point of the discussion is wrong. The whole principle of HOV/bus lanes is to encourage modal shift by minimizing the travel time of people who use buses and carpools, and the reserved lanes should discharge beyond the bottleneck point or be given priority at the bottleneck point where they have to merge with solo drivers. After a lot of heavy math, it seems that the bottom line of that paper is concerned with where the auto queues will wind up, rather than dealing with the issue of how to move more PEOPLE (not vehicles) efficiently through an entire transportation network. Of course the auto queues will be longer if they are constrained to 2 lanes (with an HOV lane beside them) instead of 3 general-purpose lanes. That's obvious, but it's missing the larger issues.

[Wouldn't you know it - my original message to UTSG was rejected for being 7 lines longer than the 250 line limit for the UTSG mailing list. If it weren't for that hideous disclaimer that my employer adds automatically ... Message to be continued ...]

All opinions are my own, not those of my employer
--------------
Colin R. Leech, P.Eng./ing.
Senior Engineer, Transit Priority
Ingénieur principal, Priorité du T.C.
City of Ottawa/Ville d'Ottawa
613-580-2424 ext./poste 13826
[log in to unmask]
--------------

This e-mail originates from the City of Ottawa e-mail 
system. Any distribution, use or copying of this e-mail or 
the information it contains by other than the intended 
recipient(s) is unauthorized. If you are not the intended 
recipient, please notify me at the telephone number shown 
above or by return e-mail and delete this communication and 
any copy immediately. Thank you.

Le présent courriel a été expédié par le système de 
courriels de la Ville d'Ottawa. Toute distribution, 
utilisation ou reproduction du courriel ou des 
renseignements qui s'y trouvent par une personne autre que 
son destinataire prévu est interdite. Si vous avez reçu le 
message par erreur, veuillez m'en aviser par téléphone (au 
numéro précité) ou par courriel, puis supprimer sans délai 
la version originale de la communication ainsi que toutes 
ses copies. Je vous remercie de votre collaboration.