Twin Cities of Minneapolis-St. Paul
Running buses on highway shoulders
The transit system in the Twin Cities of Minneapolis-St. Paul, Minnesota, USA operates an extensive express bus system, and frees it from traffic congestion with the most extensive network of transit advantages in North America.
It began in 1991 when the General Manager of Metro Transit, the Minneapolis-St. Paul transit agency, and the top officials of the Minnesota Department of Transportation (MnDOT) decided to cooperate to speed transit buses through traffic congestion. The joint effort was named Team Transit, and eventually grew to include county highway departments, city public works departments and smaller suburban transit operators. I was Metro Transit’s staff representative from 1991 until my retirement in 2006. Team Transit continues today.
The first initiative was the construction of a score of bus/HOV bypasses around ramp meters that regulated the entrance to freeways. At the time, the metre timings were very restrictive, causing long queues behind the metres and delays of up to ten minutes before cars or buses could enter the freeway. Some bypasses had been built previously, but this initiative tripled their number.
Having eliminated the delay for buses entering freeways, the next challenge was delay on the freeway itself. Traffic delays tend to be somewhat mercurial, appearing one day and disappearing the next, although some road segments are clearly worse than the others, experiencing predictable congestion every day. The volatility of traffic and wide range of travel times made it impossible to write an accurate bus schedule. One day the bus would be ten minutes late – another day it would arrive at its destination early.
I proposed running buses on the highway shoulders at reduced speed, to compensate for the narrower width and the novelty of buses operating in an unexpected location. To my surprise, MnDOT agreed to a test. It was a success and gave us the experience necessary to determine operating rules and signage requirements. Each bus shoulder lane segment is signed “Shoulder, Authorized Buses Only”. “Begin” and “End” placards are attached to the top of the sign to show where the segment begins and ends. Intermediate signs are located about every mile. If the segment is interrupted by a narrow spot, such as a bridge, a small sign showing a bus and an angled left arrow instructs the bus driver to merge back into traffic.
The great advantage of shoulder bus lanes is their minimal cost when compared with the construction of exclusive busways or High Occupancy Vehicle (HOV) lanes. Buses are heavy, so the shoulder should have the same pavement depth as the regular traffic lanes. It is also advisable to raise recessed catch basins so they are flush with the pavement, to improve ride quality. These improvements generally cost about $100,000 per lane mile, a trivial amount compared to the cost of a new busway or set of HOV lanes. The shoulders are only improved where congestion is a problem. Elsewhere, the buses run in mixed traffic, incurring no additional capital cost.
Buses only use the shoulders when traffic drops below 35 miles per hour. Once on the shoulder, the bus travels up to 15 miles per hour faster than the traffic it is passing. Bus drivers are instructed to yield to vehicles crossing the shoulder to enter or exit the highway. Should the shoulder be obstructed by a stalled vehicle, the bus merges into traffic to avoid it. The slow speeds make the use of shoulders safe. There are occasional minor sideswipes, but there have been no injury- accidents.
Shoulder lanes are not restricted to controlled access freeways. They are also installed on arterial roadways with traffic signals. When a bus approaches a signalised intersection, it passes through the right turn lane and continues straight ahead through the intersection on the same alignment as the shoulder. Sometimes this has required cutting back right turn islands to create enough width.
Experience has shown that a minimum shoulder width of ten feet is required for bus use. When travelling next to a barrier, such as a bridge railing, it is necessary to restripe the lanes to create a shoulder width of 11.5 feet, to compensate for the driver’s concern about scraping the barrier. It should be mentioned that only right-hand shoulders are used.
Since 1991, 290 miles of shoulders have been opened to buses in the Twin Cities. This has effectively freed the entire express bus network from the worst delays. While the slower speeds on the shoulder delay the bus, now there is a limit to the duration of the delay. Travel time variance is greatly reduced, thus enabling the writing of much more accurate schedules.
More important is the fact that the bus is now faster than the automobile in most corridors with shoulder use. For several years we did a public relations event pitting a regularly scheduled express bus against the hottest car we could find (Corvette, Hummer, the Batmobile). Not being delayed by traffic, the bus always won. Now it was possible to promote the bus as faster than the car, not just cheaper and safer. When we surveyed the passengers to get their reaction to the shoulder lanes, we discovered a flaw in time judgment that works in our favour. The survey (which we repeated to confirm this) showed that bus passengers overestimate both delay and time savings by a factor of two or three. They perceive delay as being two to three times longer than it actually is, and they overestimate time savings by the same amount. We would receive letters claiming that the shoulders saved 10-30 minutes per trip, when the actual savings was five minutes. Of course we did nothing to dispel this notion. Passengers told their auto-driving friends about the time savings and this word-of-mouth publicity resulted in increase in ridership.
Although ramp metre bypasses and shoulder lanes have been the primary tools to speed buses through traffic, other methods have also been employed. In a few cases, ‘bus-only’ additional turn lanes have been added at congested intersections. There are ‘bus-only’ freeway exit and entrance ramps at major park-ride lots.
Most regional express buses serve downtown Minneapolis, and for many years, slow bus speeds within downtown negated the time advantages that had been gained on the freeways. Contra-flow bus lanes had been in place for many years, but buses stopped to load passengers in every block. The number of buses exceeded the lanes’ capacity, causing further delays.
Metro Transit and the City of Minneapolis conducted a study called Access Minneapolis. Its goal was to drastically reduce travel time through downtown. The study recommended bus lanes on the “Portland model”. For some years, Portland, Oregon had operated double-width contra-flow bus lanes. Each city block along the lanes was divided into two separate bus loading zones. Buses stopped every two blocks, but only in their designated loading zone. The effect of this was to increase loading zone capacity four-fold and double bus lane throughput. Minneapolis converted its two major express bus streets, Marquette Avenue and 2nd Avenue South, to the Portland model. In addition, all bus stops display real-time bus departure information using bus GPS. The result has been a dramatic improvement in bus on-time performance and a reduction of about five minutes travel time for the one-mile trip through downtown.
In late 2010, Metro Transit opened a bus rapid transit online freeway station in the median of Interstate freeway I-35W at 46th Street in South Minneapolis. The buses are travelling a bus/HOV lane in the centre of the freeway. The station design is revolutionary because it allows buses in opposite directions to cross each other’s paths so their doors can access a centre platform. There are multiple ways to implement Bus Rapid Transit. The Twin Cities has chosen an incremental, low cost approach that arguably is the most cost effective.Aaron Isaacs Transit Service and Facilities Planner (Retd) Minneapolis, USA