A few years ago, transit managers in Troy, New York--just outside Albany--proposed improvements at some of their high-volume bus stops. Sidewalk cutouts and upgraded bus shelters were part of the plan, but they would do little to address the problem of traffic congestion, which often threw buses off schedule.
So last June, using a high-tech system called a "queue jumper," officials began giving buses priority at one particularly tangled intersection in downtown Troy. The queue jumper enables buses to move through the intersection ahead of other traffic, using a combination of pavement-embedded sensors, new and retimed traffic lights and a designated bus lane.
When a bus pulls into the stop closest to an intersection to pick up or drop off passengers, it triggers a pavement sensor. As it leaves the stop, it travels over a second sensor and into the bus travel lane, activating a specially designed traffic light. The light finishes its current cycle and then shows red in all directions, allowing the bus alone to pass through the intersection.
Engineers alleviated possible confusion at the intersection by installing a separate traffic light for the bus: In addition to the traditional red light, it includes a flashing yellow light that tells the bus driver to wait and a solid white plus sign that signals when it's safe for the bus to proceed.
As designed, the increased convenience for buses has had little adverse impact on other motorists. Because the bus-only cycle lasts less than 10 seconds and operates only when a bus approaches an intersection (15 or 20 minutes can pass between buses), other drivers may not even notice the difference. "For buses to move quicker, someone will move slower," says Carm Basile, chief of staff for the Capital District Transportation Authority, which operates Troy's buses. But "this shows that you can make transit improvements and not upset motorists."
Troy's queue-jumper experiment is just one example of the technologies that transit systems throughout the country are adopting to improve efficiency and safety, as well as to attract more riders. In some cases, the projects have been implemented quickly as a result of feedback from dissatisfied customers--while others are longer-term research efforts now coming on line.
Even as transit agencies continue developing innovative ways to enhance their service, however, they're keeping their eyes on the bottom line. In Troy, the $15,000 cost of the queue jumper was included as part of a $750,000 city intersection improvement project. Other agencies are tapping tight capital budgets for new equipment installations or relying on federal funds to bankroll more experimental technologies.
Right now, though, many are caught in a bind. The multi-year federal transportation bill that authorizes hundreds of billions of dollars expired last year and was temporarily extended only until the end of April. Transit agencies are waiting to see how much they'll receive before committing to large new expenditures. Most often, technology is incorporated when buying new vehicles that can be easily equipped, notes Lurae Stuart, of the American Public Transit Association, because retrofitting buses within several years of retirement with expensive technology often isn't worth the investment.
COLLISION AVOIDANCE
Several years ago, the Federal Transit Administration's Intelligent Vehicle Initiative steering committee identified collision-warning systems for buses as priorities to pursue using available FTA research and development dollars. Although warning systems exist for some commercial trucks, they are designed for highway driving, without the added complications of cross traffic, stoplights, pedestrians and parked cars.In California, frontal collisions (where a bus strikes another vehicle, an object or a person) accounted for about 25 percent of all transit accidents--but nearly 40 percent of the costs associated with transit accidents--during the past five years, according to Wei-Bin Zhang, who heads up the transit research program at the University of California-Berkeley.
Zhang and other UC-Berkeley researchers and engineers spent about 18 months designing a frontal-collision warning system with a team from SamTrans (San Mateo's transit agency), the California Department of Transportation and the FTA. A pilot project has been underway on three San Mateo buses for a little more than a year.
The system has three main components that collect and interpret information and issue a real-time warning to the bus driver. Sensors and cameras attached to the front of the bus transmit data that is then processed through computers installed in a box on the side of the bus. Within fractions of a second, complex algorithms programmed into the computers filter out false positives and account for whether the driver has already taken steps to ward off the danger, such as hitting the brakes or turning the wheel.
If the computer validates a danger, 18-inch-tall rectangular bars, mounted on the left and right sides of the windshield, blink to warn the driver. The lights flash with more frequency and intensify in color--from amber to bright red--as the bus nears the hazard. The intelligent system also isolates the warnings: Danger on the left or right front side of the bus triggers only the appropriate side's warning light, while danger in the middle causes both lights to flash.
Devising the ability to filter out false positives was a critical step in the project's progress, because drivers otherwise could become upset by frequent erroneous warnings or begin to ignore them altogether. "There is concern about driver overload," says APTA's Stuart.
Once the algorithms could control for the legitimacy of the warnings, designers took steps to develop a more driver-friendly warning system. The warning lights deliberately were placed in the drivers' peripheral vision, where researchers found they aren't distracting to drivers under normal conditions. "When they light up," says Frank Burton, manager of operations technology for SamTrans, "the operator instinctively reacts." Additionally, drivers can adjust the sensitivity of the system within pre-determined parameters. The onboard computers keep track of the effect those sensitivity adjustments have on the number and types of warnings issued.
About 40 drivers have been trained on the buses equipped with the warning systems; those who operate the buses most frequently have contributed feedback for improving the system, including the way the warning lights blink.
The next logical step would be to install the systems on additional buses, but a few obstacles stand in the way. For one thing, it's been challenging to prove a return on the $2.5 million investment. Since the warning systems were installed, the three buses haven't been involved in accidents, which is what would be expected statistically in any case. So researchers are relying on comparing driver behavior before and after installation of the warning system. Such analysis is underway, but Burton says that alone may not provide proof of the system's success because professional drivers are skilled at avoiding frontal collisions.
It may be tough, too, for SamTrans and its partners to find a company willing to manufacture the warning systems, something the developers say is necessary before they'll be able to install them on more buses. The means convincing a private enterprise that there's a large enough market--and demand--for the technology. Buses alone probably wouldn't justify such a venture, but the team thinks the answer lies in the system's potential to be applied to other vehicles that travel city streets, particularly small delivery trucks and vans.
In the meantime, the California researchers and engineers have teamed up with their counterparts at the Pennsylvania Department of Transportation and Carnegie Mellon University in Pittsburgh to swap technologies; a side-collision warning system was developed there. SamTrans now has added Pennsylvania's warning system, which operates in much the same fashion, to one of its buses that's already equipped with the frontal-collision warning system. The last piece of the equation, Burton says, will incorporate a rear-collision mitigation system, which will warn drivers behind the bus of potential dangers.
VIRTUAL ENGINEERS
Collision-avoidance efforts aren't focused solely on buses, either. The Illinois Department of Transportation is designing a system for trains, with funding assistance from the Association of American Railroads and the Federal Railroad Administration. The state wants to implement high-speed rail between Chicago and St. Louis, but the FRA mandates that before passenger trains can exceed 79 miles per hour in that corridor, it must be equipped with a positive train control system.Positive train control essentially acts as a backup for the train's engineer. If the system senses a hazard on or near the tracks up to 6 miles ahead of the train, it sounds an alarm and sends the engineer a visual warning via a small computer screen. The system tells the engineer what type of remedial action needs to be taken; if the engineer fails to act in time, positive train control takes over.
For example, if the train were traveling above a safe speed or approaching a stop signal, the system automatically slows or stops the train. While it's not intended to take the place of the train engineer, it can't be turned off, either. "When you have a positive train control system, it is much safer," says John Schwalbach, bureau chief of railroads for Illinois DOT. "The train cannot exceed the limits of its authority."
That action is made possible by a complex set of technologies aboard the train, along the tracks and in a mainframe computer system. To keep the cost down, some of the equipment has been purchased off the shelf; other project features have been custom-designed, including the algorithms that determine the degree of danger based on such variables as the train's speed, weight, horsepower and braking characteristics.
Schwalbach estimates that Illinois has contributed about $15 million to the project since the late 1990s; the total development cost is more than $70 million. Already Illinois DOT has run tests on parts of positive train control, including one in which a test train reached 110 miles per hour. That test was designed to check out the location- determination system, which pinpoints the train's location to within 9 feet, and to show that the project was on track.
The applications of positive train control necessary for passenger- train travel are on schedule to be completed by the end of the year, after which the FRA plans to begin extensive testing of the system. (Additional components for freight trains will continue to be installed into 2005.) Once FRA deems the corridor ready for high-speed trains, Schwalbach says Illinois gradually will begin increasing train travel beyond 79 miles per hour.
SWIPE AND RIDE
New York's Metropolitan Transit Authority made headlines last year when it officially banished tokens from its subways and buses, moving entirely to magnetic-strip fare cards. But a couple of other major cities--namely Washington, D.C., and Chicago--pioneered the adoption of smart-card technology.In a nutshell, passengers add value to plastic cards--using cash or credit--at fare machines and then swipe them over an electronic "reader" at a turnstile or fare box. Data related to the card, including its total value, entries and exits from the system, are transmitted through a computer chip, an antenna built into the turnstile and radio waves.
The cards used in both Washington and Chicago are sometimes referred to as "stored-value cards," because all the data needed to enter and leave the system is processed in real time at the gate, and passengers can see the balance remaining on the card. A backend computer system keeps track of the data, but doesn't control passenger entry and exit.
The Washington Metropolitan Area Transit Authority began using the cards on its Metrorail system about five years ago. It was a popular move. About 35 percent of subway riders use the SmarTrip card, a figure that jumps to almost half of all riders during peak hours. The success of the card on rail persuaded WMATA to expand SmarTrip to its bus system, which will eliminate fumbling for exact change, speed passenger boarding and facilitate rail-to-bus transfers.
About 100 of WMATA's fleet of 1,500 Metrobuses are equipped with SmarTrip fare boxes; the agency plans to upgrade the remainder throughout the summer. While the SmarTrip expansion is expected to make improvements to the system, it hasn't been without bugs: The implementation was originally planned for nearly two years ago but was slowed by finding--and fixing--some technological glitches.
It's not cheap, either. WMATA is spending about $23 million to make its buses SmarTrip friendly. And a plan to make SmarTrip regional by extending it to suburban Maryland and Virginia transit agencies could cost an additional $100 million, says Murray Bond, WMATA's director of SmarTrip operations.
But, Bond adds, smart cards save money, too. They lessen the use of large, expensive mechanical equipment that's maintenance prone, and if they do break, electronic machines are cheaper to fix. Electronic fare collection also cuts out the need to collect and count cash. "Right now, we spend a lot of money collecting money and dragging it around," Bond says.
The prospect of savings also attracted the Chicago Transit Authority to smart cards. It rolled out its Chicago Card, with features similar to WMATA's SmarTrip card, in November 2002, and introduced an enhanced smart card in January.
The enhanced card, called Chicago Card Plus, isn't a stored-value card--passengers can't use station vending machines or see their remaining balance as they enter and exit the system--but it does allow cardholders to access their accounts over the Internet. They also can "auto-load," meaning their credit or debit cards are automatically charged when the value of their transit smart card drops below a predetermined amount, and use the card for monthly passes.
Traditional Chicago Card users don't have monthly pass benefits-- they're automatically charged the standard $1.75 per ride--and their employer-sponsored transit dollars can't be loaded on. So far, riders have responded positively to the benefits of Chicago Card Plus: CTA sold about 21,000 in the first three months the cards were available. (The cards were free through the end of March and now cost $5 apiece.)
WMATA also charges $5 for each SmarTrip card it issues, a cost that keeps the city's bustling tourist population from using them. But the system would benefit from moving tourists through more quickly-- especially during peak tourist season or at particularly crowded stations--leading the agency to look toward the development of a smart card tourists will use. "The idea is eventually to go to a totally smart system with an economical, disposable smart card that could be used for tourists," Bond says.
