National Academies Press: OpenBook
« Previous: Chapter Three - Countermeasures to Mitigate Collisions Between Light Rail Vehicles and Motor Vehicles at Signalized Intersections
Page 23
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 23
Page 24
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 24
Page 25
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 25
Page 26
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 26
Page 27
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 27
Page 28
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 28
Page 29
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 29
Page 30
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 30
Page 31
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 31
Page 32
Suggested Citation:"Chapter Four - Case Studies." National Academies of Sciences, Engineering, and Medicine. 2008. Light Rail Vehicle Collisions with Vehicles at Signalized Intersections. Washington, DC: The National Academies Press. doi: 10.17226/14215.
×
Page 32

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

23 This chapter presents, through case studies, detailed informa- tion about the operating environment, collision history, and countermeasures tested and/or implemented for mitigating collisions between LRVs and motor vehicles at signalized intersections at a variety of transit agencies. The case studies are based on a review of the recent relevant literature and struc- tured telephone interviews with selected transit agencies. TRI-COUNTY METROPOLITAN TRANSPORTATION DISTRICT OF OREGON—MAX LIGHT RAIL TriMet began operation of their 15-mile Eastside MAX Blue light rail line in September 1986, which was among the first light rail systems in the nation. Since then, they have expanded tracks by adding 18 miles in 1998, 5.5 miles in 2001, and 5.8 miles in 2004 for a total of 44 miles. Collision History When TriMet designed the system, they assumed it would operate like a street car system; however, they realized their design of the train into the streets was different than that of a street car system. The assumption was that the “little red ball” on the traffic signal would sufficiently deter motorists from making maneuvers in conflict with LRT operations, but that assumption was incorrect. They found that left-turn drivers in particular seemed to lose attention on the traffic signal con- trolling the left-turn movement and instead seemed to be responding to other cues like the pedestrian signals and the traffic signals controlling the parallel through movement. Early on in their operations they had problems with left-turn collisions, and therefore, have made improvements that have helped control the left-turn collisions to some extent. Over the past 6 to 7 years they have experienced about a 50–50 split between left-turn collisions and right-angle collisions caused by motorists running red lights on the cross-street approaches. Countermeasures TriMet does not necessarily believe that all of the motorists violating the traffic signals are doing so deliberately, but sim- ply not paying as much attention as needed at intersections. TriMet’s goal has been to capture motorists’ attention. To do so, TriMet’s response to the left-turn problem was to duplicate the left-turn traffic signals at select locations. Left-turn traffic signals were placed on the far side of the intersection within median strips. Although a quantitative study was not per- formed, the results were not as effective as was originally hoped. Desiring to find a solution to the problem, TriMet tested the use and effectiveness of LRV-activated train- approaching warning signs, which have gone through several generations since its introduction a few years ago. The signs first started with the word “Train,” which flashed when a train approached the intersection. Next they used a flashing sign dis- playing the words, “Train Coming.” Now they use a sign with an icon of an LRV with the word “Warning.” The orientation of the LRV icon depends on the direction of the train with regard to the motorists. For left-turning motorists, the icon is the front of an LRV to indicate that the train is approaching from the same direction; the icon for the cross-street traffic is the side or profile of an LRV to indicate that it is approaching at a 90-degree angle. According to TriMet, the LRV-activated train-approaching warning signs have been effective at reduc- ing left-turn collisions. As an example of their effectiveness in Portland, in 2004 when TriMet opened the Interstate MAX Yellow Line that crosses through many signalized intersections, they established criteria for putting in the LRV-activated train-approaching warning signs. The criteria included factors such as speed, volume, school zones, crossing geometry, and sight lines. TriMet got the city of Portland to join in the effort and to use the signs liberally. As a result, they have experi- enced very few left-turn collisions since opening the MAX Yellow Line. The city of Portland has accepted the use of the LRV- activated train-approaching warning for left turns as standard practice; however, the city has not installed many of these signs for the cross-street traffic. This could in part explain the shift from a majority of left-turn collisions initially to the 50–50 split between left-turn collisions and right-angle collisions. Other effective countermeasures that TriMet has imple- mented to mitigate collisions between LRVs and motor vehi- cles at signalized intersections include • All-red phase. Washington County implemented an all- red phase that holds all traffic approaches in a stopped position while the train passes through the intersection. • Lower train speed. Along the Interstate Max Yellow Line, TriMet initially started operations with a train speed of 30 mph as opposed to 35 mph. Although they originally had plans to raise the speed to 35 mph, they have not yet seen a reason to do so. The city of Portland followed CHAPTER FOUR CASE STUDIES

suit by lowering the speed limit on the adjacent roadway to 30 mph. Since the beginning of operation, they have only experienced 11 left-turn collisions, which have been widely distributed across the many crossings along the corridor (there has not been more than one left-turn collision at any one intersection). • Signal pre-emption phasing. Initially, after an LRV pre- empted a signal, the signal would go back to the “start,” which was a green for the cross-street traffic. In the case that the protected left-turn phase was pre-empted, this phase would be skipped and motorists turning left would have to wait through another cycle. When the protected left-turn phase is skipped, some motorists may think that the signal has failed and then make the decision to violate the signal. In response, TriMet tested software that would return the signal to the phase that was pre-empted. Although this appeared to work, its use has not been well institutionalized across TriMet’s light rail system. • Increasing permissible traffic movements. Prior to the LRT line, motorists were allowed to make permissive left turns at Morrison Street in downtown Portland. Dur- ing the start-up phase of the LRT line, left turns were prohibited at Morrison Street; however, they were still allowed after an all-red phase on Yamhill Street, which runs parallel to Morrison Street. During the first year of operation, there were no collisions on Yamhill Street, despite the permissive left turns, although there were sev- eral collisions on Morrison Street. Morrison was there- fore changed to permit left turns in the same manner as Yamhill and both collisions and near-miss incidents decreased considerably (4). • Public education. Five to 6 years ago, TriMet worked to add nearly a page of language to the state driver’s manual that specifically related to driving around LRT vehicles. As a result of their efforts, TriMet has significantly reduced their collision occurrence. On average, in the 4-year period between 1994 and 1997, TriMet experienced one collision every 33,368 train-miles. (Collisions include every incident of contact, including minor fender benders, clipped mirrors, and many other incidents in which no injuries were reported and material damage was minimal.) On average, in the 4-year period between 2004 and 2007, they experienced one collision every 93,492 miles. DENVER REGIONAL TRANSPORTATION DISTRICT RTD began their LRT operation in 1994 with the 5.3-mile Central Corridor. Since then, they have expanded by adding 8 miles in 2000, 1.8 miles in 2001, and 19 miles in 2006, for a total of about 34 miles of track. RTD’s street-running opera- tions run from about 10th and Osage south of downtown Denver into and around downtown. Within the street-running section, there are approximately 35 intersections, includ- ing driveways. Along California and Stout, the LRT runs con- tra flow to two lanes of one-way automobile traffic. The track is separated by a 4-in. to 6-in. mountable curb. Downtown, 24 most of the intersections have static signs. In the 1.8-mile Cen- tral Platte Valley spur, there are a few gate-protected crossings. High Collision Locations RTD’s collision experience has been concentrated generally in a few locations. The high collision locations are discussed in more detail here. Colfax Avenue and 7th Street The intersection of Colfax and 7th Street carries high volumes of automobile traffic. Colfax is a six-lane major arterial, and 7th Street leads into the Auraria Higher Education Center. It is also a complicated intersection from a geometric perspec- tive, as 7th Street intersects Colfax on a slight curve from the north. There is a left-turn pocket lane for motorists turning left from Colfax onto 7th Street. The left turn operates with protected-permitted phasing owing to the traffic volumes at the intersection. At this intersection, RTD experiences collisions between motorists making left turns from Colfax onto 7th Street and LRVs approaching from behind, between motorists making right turns from Colfax onto 7th Street and LRVs approaching from behind, and between motorists making right turns from 7th Street onto Colfax and LRVs approaching from either direction. RTD believes that the primary reasons why colli- sions occur at this intersection are that it is a complicated, busy intersection and that motorists are either not paying enough attention or they do not understand why they are not allowed to turn. To control permissive right turns at this location, RTD used static “No Right Turn When Flashing” signs associated with a flashing yellow light (Figure 28); however, this did not work well. Drivers did not know why they were not allowed to turn FIGURE 28 No Right Turn When Flashing sign from Colfax Avenue to 7th Street in Denver.

25 when the globe was flashing, so they turned anyway and some were struck by LRVs. RTD has replaced these signs with LRV-activated turn-prohibition signs (Figure 29), which acti- vate when turns are prohibited, and these signs work better at controlling right turns than the flashing yellow. However, there are still drivers who violate the signs. RTD plans to add the LRV-activated train-approaching warning signs to provide additional information to drivers about why they are not allowed to turn. In response to right-turn motorists exhibiting risky behav- iors, including violating the active “No Right Turn” sign on the 7th Street approach at this intersection, RTD and the Univer- sity of Colorado, Denver, conducted a study (7). In an attempt to reduce risky behaviors by motorists, they implemented three treatments, which included • Extending the concrete apron 8 ft further in the right- turn lane, which created a visual contrast of the road- way surface to help approaching motorists identify the LRT–roadway crossing; • Moving the stop bar on the 7th Street approach 5 ft further upstream (from 15 ft upstream to 20 ft upstream); and. • Re-applying all pavement markings. The researchers defined several categories of risky behav- iors or “traffic violations,” including stopping 2 to 4 ft past the stop bar, stopping 6 ft past the stop bar, maneuvering before the track, stopping within 4 to 6 ft of the near rail, not stopping at the flashing no turn sign, and reversing on the tracks. A before-and-after analysis revealed a significant decrease in total risky behaviors by motorists after the treatments were installed. Speer Boulevard and Stout Street The intersection of Speer and Stout is unique. Speer is an eight-lane major arterial. The two directions are separated by a creek; thus, Speer operates like a one-way pair. Stout is a one-way street running northbound into downtown Denver. The LRT runs parallel to Stout. At this intersection (effectively two intersections), collisions occur between motorists travel- ing eastbound and westbound on Speer by running the red lights and colliding with the LRVs at a right-angle, and colli- sions between motorists making left turns from eastbound Speer onto northbound Stout. Left turns from Stout onto westbound Speer (i.e., one-way northbound to one-way westbound) and left turns from Speer onto Stout (i.e., one-way eastbound to one-way northbound) are not permitted on red, as motorists must cross the LRT tracks when turning left. Left turns are allowed only during the protected left-turn signal phase; left turns are prohibited at all other times with the use of red arrow signal displays and a sign reading “Left on Green Arrow Only.” In RTD’s experience, the red arrow signal displays work better than red balls and “No Turn on Red” signs. Motorists have more respect for the red arrow signal displays than the static signs, as they will vio- late the signs more often than the signals. Welton Corridor The Welton corridor is along the D Line north of downtown. Welton is a one-way street running northbound. There is a short section of single track that runs bi-directionally. There are a couple of intersections with traffic signals. At these locations, RTD experiences collisions between motorists approaching Welton from the east who must cross the track before entering the intersection at Welton. Motorists making right turns on red will encroach on the tracks to look to the left for a gap in traffic and then get struck by an LRV approach- ing from the right. At these intersections, there is a sign that reads, “Train Approaching When Flashing,” with an asso- ciated flashing yellow light that activates when the train is approaching. This sign has been in place since the line opened. No other countermeasures have been implemented to mitigate these collisions. Considerations During Planning Stages Based on the experience at RTD, several considerations were noted for agencies in the planning stages of LRT: • Design and engineer out the big hazards. For example, sharing the left-turn lane is confusing for motorists; they do not know when they can be in the lane and when they cannot. Even if the signage appears to be clear, it may not be. • Prohibit movements when possible. When movements are permitted sometimes and prohibited other times, motorists get mixed messages that can lead to problems. Prohibiting movements, however, can be difficult in street-running environments where the city needs to keep traffic moving. • Educate the public beforehand. Educating the public is critical, especially if there is part of the design that could FIGURE 29 LRV-activated turn prohibition signs—7th Street to Colfax Avenue in Denver.

be an issue. If a new sign will be introduced, for example, educate the public on what it means and why it is impor- tant to respect it. RTD did a public awareness campaign regarding the contra-flow section of the LRT. Using tele- vision spots, the campaign re-educated motorists to look both ways at intersections. • Take away movements with gates. In RTD’s new LRT section, they plan to remove certain movements with the use of gates. METROPOLITAN TRANSIT AUTHORITY OF HARRIS COUNTY (METRO)—HOUSTON, TEXAS In January 2004, METRO began operation of 7.5 miles of LRT, known as the Red Line, all of which are street-running. The rail line travels at grade along the surrounding streets. Most of the alignment is side-running or median-running. There is a portion of mixed-use alignment in the Texas Med- ical Center (TMC). In the TMC, motorists making left turns onto side streets and into garages and hospitals must share space with LRVs on the trackway to make their turns. Special signals and signage are used to indicate to drivers when they are allowed to be on the trackway to make turns. Left turns are not allowed on some parts of the alignment and are allowed only at signalized intersections on other parts of the alignment. Problems Initially, the most common type of collisions that occurred between motor vehicles and METRORail vehicles involved illegal left turns by motorists. In the median-running sections of the alignment, motorists would make illegal left turns at intersections where left turns were prohibited. In TMC’s mixed-use environment, drivers had difficulties understand- ing the layout and the traffic control, specifically where they had to make left turns from a left-turn pocket on the tracks. More recently, METRO has seen a shift from left-turn col- lisions to right-angle collisions resulting from motorists run- ning red lights on the cross-street approaches to signalized intersections with LRT. Mitigation Strategies In February 2004, after just 1 month of operation, METRO requested the assistance of TTI for an analysis of the rail line’s safety. A research team consisting of experts from TTI and the light rail industry reviewed the collisions and observed conditions along the rail right-of-way. After the assessment, METRO implemented a variety of countermeasures to miti- gate collisions, including improving signage for motorists. As a result, they have seen a reduction in collisions of approxi- mately 40% since the first year of operation (16). On May 31, 2007, METRO released METRORail accident statistics, which showed a continuing downward trend in the 26 number of vehicle collisions. In its first year of operation, METRORail recorded 62 accidents, 55 accidents in 2005, and 36 in 2006. As of May 31, 2007, there had been only 14 acci- dents in 2007 (17). The decline in accidents is the result of the continued implementation of the recommendations set forth by TTI. More recent changes to the system are discussed in the following sections. Active Turn Prohibition Symbol Signs with TRACK Symbol Originally at intersections where left turns were protected/ permitted, METRO used active light-emitting diode (LED) signs that displayed the words “No Turns.” Now they use active turn-prohibition symbol signs with the “Track” symbol. In-Roadway Lights Currently, they are experiencing problems with motorists running red lights with a resulting shift in collision types. METRORail now experiences as many or more right-angle collisions resulting from motorists running red lights on the cross street than those involving motorists making left-turns. In 2006, METRO began testing in-roadway flashing lights to get motorists’ attention to stop at the red lights at signalized intersections with LRT. The lights are red, installed along the stop bar, and flicker left to right at a rapid pace. Between May and October of 2007, METRO installed the in-roadway lights at 11 intersections along the Main Street cor- ridor. Prior to installation of the in-roadway lights, there were about eight red-light running accidents each year at these 11 intersections for the previous 3-year period (2004 to 2006). Since installation of the in-roadway lights at the intersections on average for about 11 months, they have experienced only two red-light running accidents at the intersections. METRO tested the in-roadway lights at two TMC inter- sections where left-turn vehicles must enter the trackway to turn left. The in-roadway lights were placed along the stop bar in the left-turn lane; however, they have not had the same success with the in-roadway lights in this application as they have had on the cross-street approaches. An FHWA study is expected to be completed in 2008 on the in-roadway lights. In addition to the modified signage in the TMC, they are currently considering the use of retractable delineators to indi- cate to motorists when they are allowed to get onto the tracks to make a left turn. In this application, the delineators would pop up along the lane line that separates LRVs and the traffic. They are currently under trial at a maintenance facility. In March 2004, all-red signal phasing was implemented at 12 signalized intersections along the alignment, and METRO reports that they have been effective in controlling the number of collisions.

27 METRO is currently planning a system expansion using the in-roadway lights and signage that have resulted in a decrease in collisions. LOS ANGELES COUNTY METROPOLITAN TRANSPORTATION AUTHORITY— METRO BLUE LINE The LACMTA Metro Blue Line opened in 1990, extending 22 miles from downtown Los Angeles to downtown Long Beach. About 6.48 miles of the Metro Blue Line is street- running, the majority of which is median-running (Figure 30), with less than 1 mile of side-running operations. There are approximately 28 crossings with gates and 62 without gates, all of which are actively controlled. The trains operate at 35 mph in the street-running sections of the alignment. Collision History In the 3 years from its July 1990 opening through late June 1993, LACMTA experienced 158 LRV–motor vehi- cle collisions at the 100 crossings on the 22-mile Blue Line. According to LACMTA, most of the collisions were caused by motorists making illegal left-hand turns into the path of moving trains, including motorists driving around gates and motorists ignoring or failing to see red “No Left Turn” signs on street-running portions of the Blue Line, where traffic signals are used instead of crossing gates. In response, LACMTA instituted its Grade Crossing Safety Improvement Program in 1992 as an effective method to discourage illegal automotive and pedestrian movements. As part of this pro- gram, LACMTA demonstrated increased police presence along the rail line; high-resolution photo enforcement sys- tems at four crossings; a wayside horn system; illuminated signage for pedestrians and motor vehicles; standardized warning devices, signs, signals, and pavement markings for LRT; and education and public awareness programs for schools, churches, community groups, and businesses along the Metro Blue Line (18). As a result primarily of the photo enforcement, LACMTA has controlled collisions at gated crossings. The issues cur- rently are LRV–motor vehicle collisions at non-gated inter- sections when the motorist makes an illegal left turn in front of a train traveling in the same direction. This type of collision makes up the majority of their collisions. Another scenario, which occurs less frequently, is the “two-train scenario.” This scenario might occur when motorists, in the left-turn pocket, see an on-coming train and think they can beat the train through the crossing. The motorists make the left turn and are struck by a train approaching from behind. Mitigation Strategies LACMTA has found all of the following strategies to be effective in mitigation collisions between LRVs and motor vehicles at signalized intersections: • Police enforcement. During a grade crossing safety improvement program initiated by LACMTA in the mid-1990s, the Los Angeles County Sheriff’s Depart- ment Transit Services Bureau established a traffic detail to provide for increased enforcement of traffic viola- tions at grade crossings along the Metro Blue Line. Sev- eral factors, including accident experience, responses to train operator surveys, and locations with broken gate arms were analyzed to determine how best to deploy the deputies. The traffic detail deputies wrote 7,760 citations in 90 days. Based on the success of this demonstration program, the MTA continued the Sheriff’s grade cross- ing safety detail for 2 years, resulting in the issuance of over 14,000 citations (19). • Photo enforcement. LACMTA began using photo enforcement in the 1990s at gated crossings in an attempt to reduce the risky behavior of motorists driving around gates. As a result, they have controlled collisions at gated crossings. In 2004, they expanded this practice to six non-gated crossings at signalized intersections. They installed nine cameras at these six intersections along with static signs in advance of the intersections that warn drivers of the photo enforcement. Drivers are cited for running red left-turn arrows whether a train is approaching or not. As a result of LACMTA’s enforce- ment efforts, accidents caused by motorists making ille- gal left turns have been reduced by 62% since left-turn enforcement began in 2004 (15). However, they still have drivers running red-turn arrows. • LRV-activated train-approaching warning signs. At major intersections, LACMTA has dedicated left- turn pockets and protected left-turn phasing. In 2000, LRV-activated train-approaching warning signs for left-turning traffic at these major intersections were added at the end of the cantilever mast arm in the Los Angeles street-running segment (as shown in Figure 30). The use of the LRV-activated train-approaching warning signs was expanded to the Long Beach street-running FIGURE 30 Los Angeles Metro Blue Line (Courtesy: Los Angeles County MTA).

segment in 2004. The LRV icon flashes when the train is approaching. • All-red signal phase. The Gold Line opened in July 2003 and runs 13.7 miles from downtown Los Angeles to Pasadena. Only about 3⁄4 of a mile of the Gold Line is street running operations. Trains operate at 20 mph and parallel very narrow streets through a residential neigh- borhood. There are seven signalized intersections, all of which have an all-red phase where the parallel traffic is also held on red as the train passes through the inter- section. LRV-activated train-approaching warning signs have also been installed for the cross-street traffic at these seven intersections (which is not done on the Blue Line). • Pavement treatment. In locations with side-running oper- ation, drivers on the cross-street approach to the inter- section encroach into the dynamic envelope of the train. To keep drivers back, they enhanced the crosswalk before the tracks to make it more noticeable. They used a colored concrete pattern, which has worked well. • Video cameras on trains. LACMTA has video cameras mounted on the windshield of the LRV pointing outward. This allows them to review collisions to determine the cause and to develop appropriate mitigation strategies. • Public outreach and education. LACMTA has main- tained a very active and aggressive public outreach pro- gram. They distribute brochures and DVDs at schools, public events, churches, and businesses. They are cur- rently planning to produce two videos: one that addresses the issue of left-turn collisions with LRVs at signalized intersections and one that addresses the issue of pedes- trian accidents with LRVs at gated crossings. They are currently waiting for funding and expect to complete the project by December 2008. LACMTA has explored a number of other counter- measures for mitigating collisions between LRVs and motor vehicles at signalized intersections. These countermeasures include • In-roadway lights. Recently, LACMTA has been looking at Houston’s use of in-roadway lights. Although Houston has installed in-roadway lights to discourage red-light running on the cross-street approaches to signalized intersections with LRT (to reduce right-angle collisions), LACMTA is considering using in-roadway lights to dis- courage motorists from running red left-turn arrows from the left-turn pocket lane. LACMTA plans to speak with the city to determine if this would be possible and, if so, to develop the specifications for the lights. • Retractable delineators. LACMTA is also awaiting the results of two different tests of retractable delineators. Houston METRO is testing retractable delineators for use in the TMC area to delineate to drivers when they are permitted to be on the tracks to make a left turn. The Michigan DOT is currently testing a type of retractable delineator to discourage drivers from driving around the crossing gates at a heavy-rail grade crossing. Again, 28 Los Angeles is considering the use of retractable delin- eators along the far “limit line” for the left-turn pocket to create a physical barrier to discourage drivers from mak- ing illegal left turns. Issues they are considering include reliability, maintenance, breakage, and vandalism, as well as issues with the city that must be explored. • Rear-view type mirrors. LACMTA looked at using rear- view type mirrors in left-turn pockets to allow left-turn drivers to view LRVs coming from behind; however, they have not used them to this point for multiple rea- sons. For example, there is an issue of where to put the mirrors. They must be in a position where the drivers can see in them. At the same time, there is the issue of the mirrors being hit by large trucks. MARYLAND TRANSIT ADMINISTRATION— CENTRAL LIGHT RAIL LINE Deployment of MUTCD Light Rail Traffic Control Devices to Improve Safety at At-Grade Light Rail Crossings in Balti- more, Maryland (2005) (20) presents the findings of a light rail traffic safety study performed for a 1.9-mile section of the Central Light Rail Line along the Howard Street corridor in Baltimore. The primary objective of the study was to examine the effectiveness of various traffic control devices and positive guidance measures to minimize the number and severity of accidents involving light rail, buses, pedestrians, and other vehicular traffic. The 1.9-mile section of the Central Light Rail Line along the Howard Street corridor was constructed within the exist- ing right-of-way on Howard Street, with two parallel tracks that run along the west side of the street for the majority of its length (Figure 31). The segment includes 17 non-gated signalized intersections. The LRT alignment includes semi- exclusive type b.3 (protected by 6-in. high curbs between crossings) and type b.4 (separated by mountable curbs, strip- ing, and/or lane designation). Originally, there was no vehi- cle traffic allowed along Howard Street; however, in 1997, northbound one-way vehicular traffic was added, resulting in a rise in the rate and number of reported accidents involving LRVs, motor vehicles, and pedestrians. FIGURE 31 Baltimore’s light rail (Courtesy: Jon Bell).

29 In 1999, a comprehensive assessment study of the corridor was performed. The study included a review of the existing traffic control devices, pedestrian crossings, risky behavior patterns, and analysis of accident data for a 5-year period from 1994 to 1998. The findings of the assessment indicated that all crossings had signs that met just the minimum requirements of the MUTCD. Other findings included • Inconsistencies in the selection and placement of advi- sory and warning signs, • Confusing advanced W10 series signs, • Turn restriction signs not incorporating the track sym- bol into the sign, • Lack of “Do Not Stop on Tracks” (R8-8) or “Do Not Drive on Tracks” (R15-6a) regulatory signs despite a continuous problem with such illegal movements, and • Worn pavement markings throughout the corridor. As a result of the assessment, MTA added many new signs and new pavement markings. No major enhancements were made to any of the left-turn signal phasing nor were barriers installed at that time. However, from 1999 to 2001, following the improvements, MTA experienced a reduction in the num- ber and rate of accidents as well as a reduction in the claims costs. Traffic and pedestrian volumes during this same time period did not change by more than 1%. In 2002, the MTA initiated several enhancement projects at the most accident prone locations. To overcome the problem of left-turn vehicles violating the left-turn signal indication, a recommendation was made to change the left-turn signal phase from a leading left to a lagging left, thus reducing the poten- tial for left-turning traffic to conflict with the LRVs. Other enhancements included replacing all green ball lenses with arrow lenses where applicable, replacing left-turn (R3-1) and right-turn (R3-2) prohibition signs with the R3-1 and R3-2 signs with track symbols, renewing all dynamic envelope and lane markings throughout the corridor with paint markings, installing new regulatory signs at unsignalized intersections, and adding treatments for pedestrians. At the time of study completion, MTA was planning to add R3-1a (no right turn across tracks) and R3-2a (no left turn across tracks) activated blank-out signs at five locations that had recurring right-turn and left-turn accidents. One noted finding of the study was the need for uniformity and consistency in the application of signs and pavement markings for controlling certain types of accidents. Specifi- cally, the delineation of the dynamic envelope proved to be a very cost-effective measure to reduce sideswipe accidents in travel sections where the travel lanes were less than 12 ft wide. In addition, the concept of a flexible barrier separation between LRVs and other vehicle traffic, although expensive, proved to be one of the most positive treatments to prohibit illegal turn- ing movements, minimize sideswipe accidents, and reduce accident severity. NEW JERSEY TRANSIT Hudson–Bergen Light Rail The New Jersey Transit’s Hudson–Bergen Line opened in April 2000 with approximately 6 miles of track. New Jersey Transit has added track in increments since the April 2000 opening. As of February 2006, the line has a total of 20 miles of track. Eight of the 20 miles operate in an exclusive right- of-way, and 12 miles operate in a non-exclusive right-of-way (Figure 32). In the street-running, mixed right-of-way in downtown Jer- sey City and other neighboring areas, the northbound track and the southbound track are separated by a median, and the south- bound track shares the one-way, one-lane travelway with auto- mobile traffic. The signals are pre-empted by the train, and the signal cycle starts as the train approaches and the cross-street traffic receives a red light. Motorists approaching intersections from the west can make only a right turn onto the one-way street. This movement is controlled with the traffic signal and a “No Turn on Red” sign. Motorists approaching intersections from the east can make only a left turn onto the one-way street, and must cross the northbound tracks to do so. This left turn is a protected-permitted movement. For the past several years, the Hudson–Bergen Line has experienced between six and nine collisions yearly at signal- ized intersections (including nine in 2007). About 90% of these collisions were right-angle collisions caused by motorists (including automobiles, buses, and tractor-trailers) running red lights on the cross streets. Only a few intersections have the LRV-activated train- approaching warning signs that flash and display the words “Light Rail.” The icon shows the profile of an LRV. A few of the intersections have a painted “block” with “crosshatch” marks. These pavement markings were created when the light rail was constructed to make the motorists aware that they are in the vicinity of the light rail line, and not in response to the right-angle collisions. Personnel from the Hudson–Bergen FIGURE 32 Hudson–Bergen alignment (Courtesy: Jon Bell).

Line visited the LRT systems in Los Angeles, Portland, and San Diego to observe what they were doing in terms of safety, and installed the LRV-activated train-approaching warning signs and the crosshatch pavement markings based on the practice in these cities. New Jersey Transit–Newark Light Rail New Jersey Transit’s Newark Light Rail is a 6.5-mile light rail line that operates as a rapid transit link between a terminal sta- tion at Newark Broad Street Station through Newark Penn Sta- tion located in Newark (Pennsylvania Railroad Station or Penn Station–Newark) to the Grove Street Station in Bloom- field, New Jersey. The line is in an underground tunnel for 1.7 miles either at grade or depressed cut for 3.8 miles, and includes approximately 1 mile of street-running territory. The street-running territory is part of the Broad Street Extension. The extension to Broad Street Station is mixed street-running territory where automobile traffic operates parallel to the guideway, which is separated from the traffic lanes by a 6-in.- mountable granite curb. The street-running extension align- ment has 14 grade and pedestrian crossings. These intersec- tions are protected by traffic control devices that are integrated with the train control system to give priority to the LRVs and to prevent conflicting signals and unsafe vehicular move- ments. Each intersection is marked in accordance with the provisions of the MUTCD and New Jersey DOT diagnostic team recommendations. Newark Light Rail has experienced between about two and four collisions annually at the signalized intersections. In 2007, there were two such incidents: a right-angle collision caused by motorist running a red light and one involving a motorist within the dynamic envelope of the train. SACRAMENTO REGIONAL TRANSIT DISTRICT The Sacramento RT operates approximately 37 miles of light rail, which links the eastern and northeastern suburbs with downtown and South Sacramento. Approximately 29 miles of the light rail system are double track, with the remaining being single track. Sacramento RT began light rail service in 1987, expanding their system in the late 1990s, and continuing with expansions that nearly doubled their system during the 2000s. The Sacramento RT light rail system is experiencing particu- lar issues with each of its operating environments, including side-running, shared-lane, and a pedestrian mall. Collision Experience Side-Running Environment In one section of 12th Street, which runs one-way southbound, the light rail operates in a side-running environment on the east side of 12th Street. Therefore, the cross-street traffic approach- ing from the east must cross the tracks before entering the 30 intersection. The intersection of 12th Street and Ahern Street is controlled by a stop sign only on the Ahern approach. Although this is an unsignalized intersection, there is a pedes- trian head that displays the words “No Left Turn” when a train is approaching (Figure 33). At this intersection, some drivers pull up past the stop bar, look right for a gap in traffic, and are struck by the train coming from their left. Two primary issues are contributing to this problem. First, drivers cannot see the pedestrian head from the stop bar. Second, the pedestrian head is not timed properly with the approach of a train. Sacramento RT has been working with the city and the engineers, and they plan to take the following steps to reduce collisions when funding becomes available: • Add striping and pavement markings on the cross-street approach to cover approximately 15 ft from the nearest rail to delineate a zone where drivers should not be stopped. • Reposition the signal head so that drivers can see it from the stop bar. • Make signal timing adjustments to the “No Left Turn” sign to give motorists enough advance warning to make decisions. • Request funding to replace current pedestrian heads that display “No Left Turn” with the activated no left- turn symbol blank-out sign. Mixed-Use Operations Sacramento RT has mixed-use operations along parts of 12th Street, which is a one-way street running into downtown Sacramento. It carries four lanes of automobile traffic, drop- ping lanes one-by-one into downtown, where there are two lanes. The southbound tracks share the eastern-most lane with automobile traffic, whereas the northbound train operates in its own right-of-way (Figure 34). Therefore, drivers make left turns in a shared lane with the train and turn across the tracks carrying trains in the opposing directions (Figure 35). The problem that Sacramento RT is currently experiencing is that FIGURE 33 Ahern Street approach to 12th Street with pedestrian head train approaching warning sign (Courtesy: Sacramento Regional Transit District).

31 drivers making left turns from the “#2 lane” (instead of mak- ing left turns from the lane shared with the train) are being struck by the train in the adjacent lane traveling in the same direction. There is no signage indicating lane assignments and permitted movements; however, there are pavement markings (arrows) in each lane that indicate the permitted movements. Possible issues with the pavement markings are that they have faded and, when vehicles are stopped at the light, drivers might not be able to see them. Sacramento RT has requested that the city add lane-use signs to indicate lanes and permitted movements. There is a flashing yellow associated with the outbound train, and they have not had a problem with colli- sions between left-turning motorists and LRVs traveling in the opposite direction. Pedestrian Mall Another location where they are experiencing problems is downtown at the intersections of 9th and 10th and O Streets. Because there is a pedestrian mall along O Street; motorists traveling along 9th and 10th streets are not allowed to make turns onto it. The traffic on 9th and 10th is controlled by traf- fic lights that turn red for pedestrians and for trains. They are experiencing drivers running the red lights at O Street. Gen- erally, the drivers stop at the lights to look for pedestrians; however, when they do not see any pedestrians, some decide to run the light, not realizing a train is coming. The collision is a right-angle collision between the LRVs and the motor vehicles. In rare cases, a motorist does not stop at all and gets struck by the train in the intersection. Sacramento RT hypothesized that this may be a result of the traffic signal progression provided along 9th and 10th Streets. In other words, drivers may be expecting the lights at O Street to turn green as they are approaching; however, if the train has pre- empted the signal, the light will not be green as expected; therefore, motorists run the light and a collision occurs. Although they cannot confirm this, Sacramento RT suspect that this is a factor in at least some of the collisions. DALLAS AREA RAPID TRANSIT DART operates more than 40 miles of light rail, most of which is located in exclusive right-of-way. Motor vehicles and LRVs are, however, controlled exclusively by traffic signals in two different environments, a median-running segment of the South Oak Cliff Line and the central business district Transit- way Mall in downtown Dallas. The segment of the South Oak Cliff Line that is controlled by traffic signals is approximately 2-miles long and runs in the middle of a four-lane divided arte- rial (median-running). The LRT track is separated from the motor vehicle travel lanes by a raised barrier curb. There are 12 at-grade crossings in the 2-mile segment. This section was part of a system expansion in 1997. After the expansion, DART began experiencing a number of collisions between LRVs and motorists making illegal left turns across the tracks from the protected left turn. Although lead-lag left-turn opera- tions are normally preferred to improve progression along the arterial, dual left-turn phasing was implemented to improve compliance with the left-turn restriction. Even with these FIGURE 34 12th Street mixed-use right-of-way in Sacramento (Courtesy: Sacramento Regional Transit District). FIGURE 35 12th Street and E Street: Vehicle occupying shared lane (left) and LRV occupying shared lane (right) (Courtesy: Sacramento Regional Transit District).

traffic controls in place, left-turn violations continued to occur (21). In an effort to mitigate these left-turn violations and collisions, in 1999 DART implemented “train coming” signs at the intersections. These signs display an LRV icon and the text, “Train Coming.” The signs illuminate on detection of an LRV from either direction. The signs are placed in the median on a pedestal pole directly across from the left-turn pocket lane. There is also an additional left-turn signal in this location (21). Since the installation of these signs, the number of collisions between LRVs and left-turning motorists has been reduced dramatically. VALLEY METRO (PHOENIX) The 20-mile initial METRO light rail line is scheduled to begin operations in December 2008. The METRO system will oper- ate at street level in a lane separated from traffic, and trains will travel primarily in the street median. When METRO construc- tion is complete, there will be improved light-rail synchro- nized signals at 148 intersections. That number includes 15 new signals added to create more U-turn areas for business access (22). METRO is currently testing LRVs on Washington Street between 48th and 56th Streets. During this testing, auto- mobile traffic travels alongside the LRV that is operated at very slow speeds (22). METRO is working with residents, schools, and businesses in the area to ensure that they all understand the basics of light rail safety. Beginning December 3, 2007, METRO stopped using police officers to guide traffic at inter- sections. Instead, both LRVs and automobile traffic are being controlled by METRO’s traffic management system. The new system will control all intersections on Washington Street between 44th Street and Priest Drive (23). In the spring of 2008, METRO gradually began expanding its vehicle testing activities. At that time, they began a public education campaign on the safety rules for light rail with the goal of raising the awareness of rail safety and encouraging safe behaviors. METRO plans to use the communication resources of their partner cities and sought communication partnerships with the Valley’s businesses, neighborhood groups, community organizations, and the news media (24). To make it safer and more street-friendly, METRO’s light rail design was influenced by discussions with other rail authorities and cities. Observing and researching similar light rail systems around the country proved to be invaluable in determining appropriate system enhancements for the 32 METRO system. To improve traffic operations and to min- imize common types of collisions between LRVs and motor vehicles, the following features were incorporated into the planning, design, and/or construction of the METRO light rail system (24): • Protected left- and right-turn lanes. Turns across the tracks will be made only from exclusive (left- or right- turn-only) turn lanes. “Protected” signals will control left- and right-turn movements by red, amber, and green arrows, which are considered to be the safest form of turning control used by traffic engineers. Other cities tried using special “No Left-Turn” or “No Right-Turn” signs in shared lanes that activate when a train is approaching; however, based on discussions with the LRT operating systems, these signs were mostly ignored by motorists, resulting in accidents. • Longer left-turn lanes. Left-turn storage bays will be lengthened to handle projected 2020 traffic condi- tions, including storage for the added U-turns that will be required to access some driveways and local streets. Adequate storage is critical to improving safety and reducing congestion caused by traffic backing into through-travel lanes. • LRV cameras. Cameras will be installed on the Metro LRVs so that train operators can better see obscured pedestrians and obstructions. They will also be installed on the vehicle exterior for monitoring and recording traf- fic conditions, unsafe driving behaviors, and accidents. • Controlled track crossings. For safety reasons, traffic will be allowed to cross the tracks only at a controlled loca- tion. Green-arrow signal indications for left turns and U-turns will replace solid-green balls. Special signing, such as the flashing LRV-activated train-approaching sign, will be installed. • Six-in. curbs. METRO will use 6-in. curbs to separate and protect traffic from the rail guideway. Some cities use curbs, and others use concrete barriers (e.g., San Jose), paint (e.g., Salt Lake City), or traffic buttons (e.g., Houston) to delineate the dynamic envelope of the train. • New frontage roads. Access on one-way streets will be maintained for businesses with the use of new 16-ft frontage roads when needed. Drivers will be able to enter the frontage roads at traffic signals and exit at signalized slip ramps to re-enter the main flow of traffic. New spe- cially designed frontage roads were designed to handle large trucks and emergency access and are necessary to maintain safe business access on the one-way streets. This is the first design of its kind being used specifically for light rail applications in the United States.

Next: Chapter Five - Conclusions »
Light Rail Vehicle Collisions with Vehicles at Signalized Intersections Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's Transit Cooperative Research Program (TCRP) Synthesis 79: Light Rail Vehicle Collisions with Vehicles at Signalized Intersections explores mitigation methods tested and used by transit agencies to reduce collisions between light rail vehicles (LRVs) and motor vehicles where light rail transit (LRT) runs through or adjacent to highway intersections controlled by conventional traffic signals.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!