National Academies Press: OpenBook

Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH (2012)

Chapter: Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices

« Previous: Chapter 2: Identification of Participating LRT Agency and Potential Study Crossings
Page 22
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 22
Page 23
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 23
Page 24
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 24
Page 25
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 25
Page 26
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 26
Page 27
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 27
Page 28
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 28
Page 29
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 29
Page 30
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 30
Page 31
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 31
Page 32
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 32
Page 33
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 33
Page 34
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 34
Page 35
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 35
Page 36
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 36
Page 37
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 37
Page 38
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 38
Page 39
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 39
Page 40
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 40
Page 41
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 41
Page 42
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 42
Page 43
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 43
Page 44
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 44
Page 45
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 45
Page 46
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 46
Page 47
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 47
Page 48
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 48
Page 49
Suggested Citation:"Chapter 3: Assessment of the Operational and Safety Aspects of Existing, Alternative, and Possible Supplemental Traffic Control Devices." National Academies of Sciences, Engineering, and Medicine. 2012. Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH. Washington, DC: The National Academies Press. doi: 10.17226/22822.
×
Page 49

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.

TCRP Web-Only Document 53 18 CHAPTER 3: ASSESSMENT OF THE OPERATIONAL AND SAFETY ASPECTS OF EXISTING, ALTERNATIVE, AND POSSIBLE SUPPLEMENTAL TRAFFIC CONTROL DEVICES The next step in the project was to analyze the operational and safety aspects of existing traffic control devices, as well as alternative and possible supplemental traffic control devices that could be used at the intersections identified for study, namely:  North First Street and Brokaw Rd.;  North First Street and Charcot Ave.; and  North First Street and Trimble Rd. The general approach to this analysis was to gather and analyze a variety of data that would indicate how well the intersections currently operate and the existing level of safety at the intersections. Specifically, the approach to the analysis was three-fold as is presented in Table 7. Table 7. Approach to the Assessment of Existing, Alternative, and Possible Supplemental Traffic Control Devices Approach For the Assessment of: Existing Traffic Control Devices Alternative and Possible Supplemental Traffic Control Deices (1) Gather and analyze historical safety-related data provided by the local agencies.  (2) Observe and analyze operations and driver behavior through field video data collection.  (3) Examine qualitative data collected through focus groups with local drivers.   The data and results are presented below. ASSESSMENT OF THE OPERATIONAL AND SAFETY ASPECTS OF EXISTING TRAFFIC CONTROL DEVICES The primary means for assessing the operational and safety aspects of existing traffic control devices were (a) to gather and analyze historical safety-related data and (b) observe and analyze operations and driver behavior through field video data collection. This section of the report presents a synthesis of the data and findings from the analyses. Qualitative data collected through focus groups with local drivers was also examined, and these results are presented in the subsequent section. Analysis of Safety-Related Data The team requested and examined the following safety-related data:  Historical crash data; and  Historical near miss data. These data are presented for the three study locations below.

TCRP Web-Only Document 53 19 Historical Crash Data The research team gathered crash data from both the City of San Jose and the VTA. Both data sets represented 3 years worth of crashes. The city reported motor vehicle crashes that occurred between May 10, 2006 and May 10, 2009. The VTA reported LRV-motor vehicle crashes that occurred between May 19, 2006 and April 22, 2009. Tables 8 and 9 present a high-level summary of the city-reported motor vehicle crashes. Table 8 shows the total number of motor vehicle crashes that occurred at each of the study intersections along with the total number of fatal crashes and total number of injury crashes. Table 9 highlights the crash types of particular interest when considering LRT operation through an intersection: those crashes involving pedestrians, bicycles, and LRVs, as well as those crashes involving red-light running or right-angle crashes. The data shown in Table 9 are not meant to provide a complete categorization of the crashes but rather to point out the types of motor vehicle crashes that are potentially of higher risk at intersections with LRT. In all, there were 109 reported motor vehicle crashes at the three intersections over the 3-year period. There were no fatalities, and less than one-third of the crashes involved minor injuries. Six crashes (6 percent) involved LRVs, and four crashes (4 percent) involved pedestrians. A total of 17 crashes (16 percent) were a result of vehicles running red lights. Table 8. Total, Fatal, and Injury City-Reported Motor Vehicle Crashes at Study Intersections Intersection of North First St. @.. Total Crashes Total Fatal Crashes Total Injury Crashes Brokaw Rd. 64 0 15 Charcot Ave. 14 0 7 Trimble Rd. 31 0 10 Total 109 0 32 Source: City of San Jose Department of Streets and Traffic. Table 9. Critical Crashes at Study Intersections Intersection of North First St. @.. Number of Crashes Involving: Ped Bike LRV Vehicle Running Red Light Right Angle Crash Brokaw Rd. 2 2 4 9 9 Charcot Ave. 1 1 1 4 5 Trimble Rd. 1 1 1 4 5 Total 4 4 6 17 19 Source: City of San Jose Department of Streets and Traffic. Appendix A presents a more detailed breakdown of the city-reported crashes in terms of a number of factors, including crash type, probable cause, direction of movements, day of week,

TCRP Web-Only Document 53 20 time of day, weather conditions, lighting conditions, and surface conditions at the three intersections. The following discussion summarizes the issues identified as being most relevant to this project:  Intersection of North First Street and Brokaw Rd.—During the 3-year period, there were 64 reported crashes at this intersection, by far the highest crash rate of the three study intersections; however, there were no fatalities, and only 23 percent of the crashes resulted in injuries. A large majority of crashes (over 80 percent) occurred during daylight hours, clear conditions, and on dry pavement. Rear-end crashes were the predominate type of crash (39 percent). Right-angle crashes accounted for about 14 percent of the crashes. Five of the right-angle crashes were between southbound and westbound vehicles, and four were between southbound and eastbound vehicles. In four of the nine right-angle crashes, red-light running was cited as the cause of the crash, and in one of these cases it was the cross-street vehicle that was cited for the red-light violation. In the 3 years, there were four LRV-motor vehicle crashes. Two of these were a result of a northbound left-turn violation, one involved unsafe backing, and one involved an unsafe lane change. The two left-turn related LRV crashes resulted in injuries; three people suffered minor injuries in these crashes. While there was a large number of crashes at this intersection over the past 3 years, red- light running on North First St. (in particular, the southbound through movement and the northbound left-turn movement) appears to be the biggest concern at this intersection.  Intersection of North First Street and Charcot Ave.— Over the 3-year period, there were 14 reported crashes at this intersection, the lowest crash rate of the three study intersections. There were no fatalities, and eight people were reported to have been injured in seven of the crashes. The predominant crash type was right-angle crashes (36 percent). Of the five right-angle crashes, two drivers were cited for running a red light (one southbound on North First St. and one eastbound on Charcot Ave.) In two of the crashes the cause was “unknown.” In the other, a driver on Charcot Ave. made a right turn on red and was hit by a northbound vehicle. In the 3 years, there was one LRV-motor vehicle crash. This crash was a result of a northbound left-turn violation. One person suffered a minor injury. There is not a clear crash pattern at Charcot Ave. that would indicate a particular operational or safety concern or problem. In fact, the overall low crash rate at this intersection, as compared to the others, indicates that it is the safest of the three intersections.  Intersection of North First Street and Trimble Rd.— Over the 3-year period, there were 31 reported crashes at this intersection. None of the crashes resulted in a fatality, and injuries were reported in about one-third of the crashes. As with the other intersections, the large majority of the crashes occurred during daylight hours and under clear, dry conditions. Five (16 percent) of the crashes at Trimble Rd. were classified as right-angle crashes. Three of the right-angle crashes were a result of red light running (two on North First St. and one on Trimble Rd.). In the other two crashes the cause was “unknown.” In the 3-year period, there was one LRV-motor vehicle crash. This crash involved a motor vehicle making a southbound U-turn. The cause for this crash was recorded as “unknown.” As with the other intersections, there is not a clear crash pattern at Trimble Rd. that would indicate a particular operational or safety concern or problem.

TCRP Web-Only Document 53 21 For this project, it is important to characterize and assess the LRV-related crashes at the three study intersections. Table 10 characterizes the six LRV-related crashes, as reported by the City of San Jose Department of Streets and Traffic. Table 10. Summary of Crashes at the Study Intersections Involving LRVs Intersection of North First St. @ To ta l C ra sh es To ta l L RV F at al Cr as he s To ta l L RV In ju ry C ra sh es Number of Crashes Involving LRV and: Ped Bike Left-turn Vehicle Running Red Light from North First St. Vehicle Making Unsafe Backing Maneuver Vehicle Making Unsafe Lane Change Unknown Brokaw Rd. 4 0 0 0 0 2 1 1 0 Charcot Ave. 1 0 0 0 0 1 0 0 0 Trimble Rd. 1 0 0 0 0 0 0 0 1 Source: Data were provided by the City of San Jose Department of Streets and Traffic. As stated previously, the city data show six LRV-related crashes in the past 3 years at the three study intersections. Four of the six crashes (67 percent) occurred at Brokaw Rd. One LRV- related crash was reported at each of the other two intersections in the past 3 years. Of the LRV- related crashes, none involved pedestrians or bicyclists. One-half of the crashes (two of the four) involved vehicles making left turns from North First St. onto the cross-streets. The team also received crash data from the VTA. Table 11 characterizes the LRV-related crashes as reported by the VTA. It is not surprising that the data obtained from the VTA are not completely consistent with the city-reported data, because crashes that occur within the track right-of-way may be reported solely by the VTA. In addition, the VTA data are somewhat more descriptive of the LRV-motor vehicle collisions than the city-reported data. The VTA data show 11 total LRV-related crashes over the 3-year period (as opposed to 6 crashes in the City-reported data). Six of the 11 (55 percent) crashes occurred at Brokaw Rd. There were two LRV-related crashes reported at Charcot Ave. and three LRV-related crashes reported at Trimble Rd. in the past 3 years. Of the LRV-related crashes reported by the VTA, none involved pedestrians or bicyclists. Nearly all of the crashes (10 of the 11 crashes, or 91 percent) involved vehicles making left turns from North First St. onto the cross-streets.

TCRP Web-Only Document 53 22 Table 11. Summary of Crashes Involving LRVs within VTA Right-of-Way Intersection of North First St. @ To ta l C ra sh es To ta l L RV F at al Cr as he s To ta l L RV In ju ry Cr as he s Number of Crashes involving LRV and: Ped Bike Left-Turn Vehicle Running Red Light from North First St. Left-Turn Vehicle Violation LRV Signal Violation Brokaw Rd. 6 0 2 0 0 5 1 0 Charcot Ave. 2 0 1 0 0 1 1 0 Trimble Rd. 3 0 0 0 0 1 1 1 Source: Data were provided by the VTA. Historical Near Miss Data In addition to the actual crash data, the VTA provided information on near-miss incidents. These data are generated through reports filed by the LRV operators. The reports are based on specific actions taken by the operators, such as applying emergency brakes or slowing down to avoid a vehicle making an illegal movement, pedestrian(s) standing on the tracks, vehicles stopped on tracks, etc. While they do not represent actual crashes, these data provide excellent surrogate safety measures by identifying high-risk behaviors that could potentially lead to actual crashes. Table 12 categorizes the near-miss incidents reported at the three study intersections for the same 3-year period from 2006 to 2009. The categories shown in the table were taken from the VTA data. It is important to note that the categories are not mutually exclusive, as some crashes are reported in multiple categories. In all, there were 29 near-miss incidents reported by train operators at the three study intersections during the 3-year period. Seventeen (57 percent) of these were reported at Brokaw Rd., while six (21 percent) were reported at each of the other two intersections. A majority of the near-miss incidents (21, or 72 percent) involved left- or U-turning vehicles, many of which were classified as a “left-turn violation.” There were two near-miss incidents, both at Brokaw Rd., involving a pedestrian crossing in front of the train. Twelve (41 percent) of the near-miss incidents involved the LRV operator applying the maximum brake, and an additional two (7 percent) involved the LRV operator applying the Level 5 brake (non-emergency service braking).

TCRP Web-Only Document 53 23 Table 12. Summary of Reported Light Rail Near Miss Incidents (5/22/06 to 4/21/2009) at Study Intersections on North First Street Intersection of North First St. @ To ta l R ep or te d Ne ar -m iss In cid en ts Number of Near-miss Incidents Involving: Le ft- tu rn in g Ve hi cle (i nc lu di ng U- tu rn s) Ve hi cle s w ith L ef t-t ur n Vi ol at io n Re d- Li gh t R un ni ng V eh icl es (D ire ct io n Un kn ow n) Pe de st ria n Cr os sin g in fr on t o f LR V Bi ke C ro ss in g in fr on t o f L RV Tr ac k I nt ru sio n LR V Ap pl yin g Ma xim um B ra ke (M B) LR V Ap pl yin g Le ve l 5 B ra ke (B 5) Brokaw Rd. 17 14 7 2 2 0 0 5 1 Charcot Ave. 6 3 2 5 0 0 0 4 0 Trimble Rd. 6 4 0 0 0 1 1 3 1 Source: Data were provided by the VTA Summary of Historical Safety-Related Data Motor vehicle crash experience at the three study intersections included a range of crash types, a range of vehicle movements (i.e., vehicle direction of travel and movement at intersections), and a range of probable causes. As a result, it is difficult to identify specific motor vehicle crash patterns in the data that would be suggestive of any particular operational or safety problems or concerns, particularly as they related to LRT. Two discernable patterns were identified. First, at the intersection of North First St. and Brokaw Rd., southbound vehicles were involved in all nine of the right-angle crashes, and the drivers appear to have been at fault in at least five of the crashes (three were cited for running a red-light, one was hit while making a right turn on red, and one left the scene of the crash) and possibly in as many as seven (in two of the crashes, no fault or violation was identified). Southbound vehicles also appear to be over-represented in right-angle motor vehicle crashes at Trimble Rd., although in only one case was the driver cited for running a red light (the other two are unknown). However, with regard to the LRT, this is not a major concern. The other issue of concern identified in the crash data, and one that is of more concern as it relates to LRT, is the left-turn violations at all three intersections. There were 11 LRV-motor vehicle crashes over the 3-year period at the 3 study intersections, 6 of which occurred at Brokaw Rd. Nearly all of these crashes (10 of the 11 crashes) involved vehicles making left turns from North First St. onto the cross-streets. In addition to the 11 crashes, there were 29 left- turn near-miss incidents reported by train operators, about 60 percent of which occurred at Brokaw Road. As with the crash data, left-turn violations make up the majority of the near-miss incidents. This pattern of left-turn violations is of obvious concern as it relates to LRT. This is the primary reason for VTA’s ongoing left-turn improvement project. It is encouraging that in only three of the 19 right-angle crashes reported in the 3-year period (one at each of the three intersections) was the cross-street driver cited for running a red light, a

TCRP Web-Only Document 53 24 potential high-risk incident with regard to the LRT. Red-light running on the cross-street was also not something predominate in the near-miss data. Observation of Operations and Driver Behaviors (Field Video Data) In addition to the historical safety data, the team conducted field observational studies as a means of assessing the operational and safety aspects of the existing traffic control devices. This section of the report summarizes the methodology for the field video data collection and the results of the analysis. Field Video Data Collection Methodology Cameras were set up at approximately 200-250 feet in advance of the curb line of the intersection. This allowed the camera to capture all vehicular, pedestrian and light rail movements on the North First St. approaches, as well as observing the cross-street movements. At Brokaw Rd. and Trimble Rd., cameras were set up to on the northbound approaches to the intersections. At Charcot Ave., cameras were set up on both the northbound and southbound approaches to the intersection. Cameras were also positioned to capture the through and left-turn signal indications. Figures 7 through 9 show the placement, orientation, and capture area of the cameras at Brokaw Rd., Charcot Ave., and Trimble Rd., respectively. Figure 7. North First Street at Brokaw Road

TCRP Web-Only Document 53 25 Figure 8. North First Street at Charcot Avenue Figure 9. North First Street at Trimble Road

TCRP Web-Only Document 53 26 A group of five technicians was trained off-site and in the field as camera operators and observers before the actual field data collection commenced. Training included operating the video equipment; positioning the cameras and video settings to maximize the coverage areas, conflict events, and risky behavior maneuvers that should be identified; making notes of the implications of any of these conflicts or risky behaviors, if any; and manual recording of any special events such as crashes and conflicts that could not be seen within the camera view or that could bias the risky behavior observations. Following these training sessions, an engineer supervisor visited each intersection to set up the cameras and mark the position of the tripod legs on the sidewalk so that identical set-ups could be repeated on each of the five data collection days. The angle of the camera view and optimal zoom ratio to capture the intersection area also were noted at each location. Camera operators were also instructed to be inconspicuous but able to observe and time-stamp written notes of special events such as crashes, trapped pedestrians, near misses, or track intrusions. This information would later assist the data reduction team to observe and confirm those events. Observations were recorded on five consecutive weekdays starting on Monday, June 22, 2009 between the hours of 7:00 a.m. and 9:00 a.m., 11:00 a.m. and 2:00 p.m., and 4:00 p.m. and 7:00 p.m., for a total of 8 hours each day. In total, 160 hours were recorded. A typical headway between LRVs in the same direction was 7.5 to 8 minutes, which resulted in 64 light rail crossings in each direction (128 crossings in both directions at each intersection on each observed day). LRV presence and headways were recorded by the observers throughout the 8- hour data collection period. Based on this headway information, there were 1,920 crossings where an LRV was present in either direction during the entire 160-hr data collection period at the three study intersections. Collected data were reduced and summarized in the office by two trained technicians. The technicians tallied the observations for a pre-determined set of risky behaviors. Both technicians were supervised by the same engineering supervisor that trained and supervised the field technicians. Risky behavior data that were collected from the video observations included the following categories:  Motorists entering the intersection on change/clearance intervals (mainline or cross-street);  Motorists entering the intersection on red interval (mainline or cross-street);  Hesitation of left-turners and/or stopping beyond the stop line on North First St.;  Unsafe lane change and/or backing for motorists on North First St.;  Motorists stopping on tracks or queuing on tracks (due to congestion, etc.) for North First St. mainline left-turn or cross-street traffic movements;  Pedestrians standing on or between tracks;  Pedestrians/bicyclists crossing intersection on Don’t Walk display;  Bicyclists stopping on tracks; and  Light rail operator entering and clearing intersection during horizontal bar. Field Video Data Results Baseline field video data of drivers’ and pedestrians’ risky behaviors were collected during the week of June 22, 2009. Data were reduced by tallying the observations regarding a pre-

TCRP Web-Only Document 53 27 determined set of risky behaviors. In order to clarify risky driver behavior patterns, the following were developed to categorize relevant risky behavior characteristics:  Group 1: Left-Turn and Cross-Street Movements;  Group 2: Right-of-Way (ROW) and Positioning Related;  Group 3: Stopping on Tracks;  Group 4: Pedestrian or Bicycle Related;  Group 5: Light Rail Vehicle Related; and  Group 6: Risky Behavior Observed without Direct LRT Impact. Table 13 shows the categories and descriptions of risky behavior data collected in the field, and Table 14 summarizes the results of the risky behavior observations recorded for a total of 160 hours at three intersections locations. Table 13. Risky Behavior Categories and Descriptions RISKY BEHAVIOR OBSERVATION DESCRIPTION OF BEHAVIOR GROUP 1: LEFT-TURN AND CROSS-STREET MOVEMENTS Mainline left-turn change and clearance interval violation ............................................................... Left turn motorist from North First St. enters intersection at the end of the yellow change interval or during the all-red clearance interval with and without train presence Mainline left-turn red-light violation ..................... Left turn motorist from North First St. Motorists enters intersection during the red interval with and without train presence Mainline U-turn change and clearance interval violation ............................................................... U-turn motorist from North First St. enters intersection at the end of the yellow change interval or during the all-red clearance interval with and without train presence Mainline U-turn red-light violation ....................... U-turn motorist from North First St. enters intersection during the red interval with and without train presence Cross-street red-light violation ............................ Motorist on cross-street enters intersection during the red interval with and without train presence GROUP 2: RIGHT-OF-WAY AND POSITIONING RELATED Mainline through lane stop bar intrusion ............. Through motorist on North First St. stops on or passed the stop bar during the red interval Mainline left-turn stop bar intrusion ..................... Left- or U-turn motorist on North First St. stops on or passed the stop bar during the red interval Lane change violation ......................................... Motorist in either the through or right-turn lane makes an illegal lane change to turn left from North First St. Track intrusion violation ...................................... Motorist wrongly enters track right-of-way GROUP 3: STOPPING ON TRACKS Vehicle stopped on tracks ................................... Motorist stops on tracks for reasons other than queuing, blocking, or yielding to violating motorists or pedestrians

TCRP Web-Only Document 53 28 RISKY BEHAVIOR OBSERVATION DESCRIPTION OF BEHAVIOR Mainline left-turn vehicles queued on tracks ....... Left turn motorist from North First St. stops on tracks due to a queue spillback on cross-street Cross-street left-turn vehicles queued on tracks Left turn motorist from cross-street stops on tracks due to a queue spillback on North First St. Cross-street through vehicles queued on tracks Through motorist from cross-street stops on tracks due to a queue spillback in the downstream receiving lanes GROUP 4: PEDESTRIAN OR BICYCLE RELATED Pedestrian standing on tracks ............................ Pedestrian did not complete crossing during Walk or Flashing Don’t Walk signal display and is standing between the tracks or in the track right-of-way envelope Pedestrian intersection violation ......................... Pedestrian crosses North First St. during the Don’t Walk signal display Pedestrian jay-walking violation .......................... Pedestrian crosses North First St. outside of a designated crosswalk Bicyclist mainline left-turn red-light violation ....... Left turn bicyclist on North First St. enters intersection during the red interval Bicyclist cross-street red-light violation ............... Bicyclist on cross-street enters intersection during the red interval GROUP 5: LIGHT RAIL VEHICLE RELATED Light rail vehicle violation ................................... Light rail vehicle enters or clears intersection during the horizontal white bar LRV signal indication Light rail emergency braking ............................... Light rail operator applies emergency/maximum brakes GROUP 6: RISKY BEHAVIOR OBSERVED WITHOUT DIRECT LRT IMPACT Other ................................................................... Any event that is not described above and considered 'risky', but does not have a direct impact on the LRT operations and/or ROW (e.g., vehicles backing up to change lanes after stopping at the intersection, right turn maneuvers from the left turn lane), any signs of aggressive driver behavior (i.e., cutting off, abrupt stops and lane change maneuvers, etc.), and right turns on red that conflicts with opposing traffic. Vehicles backing up to change lanes ................. Right turn or through maneuvers from the left lane ..................................................................... Left turn from through or right turn lane .............. Aggressive driver behavior (cutting off, abrupt stops) ................................................................. Right turn on red that opposes through traffic .... Mainline through red-light violation ..................... Through motorist from North First St. Motorists enters intersection during the red interval without train presence

TCRP Web-Only Document 53 29 Table 14. Summary of Risky Behavior Observations RISKY BEHAVIOR OBSERVATION Number of Risky Behavior Observations North First St. at Brokaw Rd. Charcot Ave. Charcot Ave. Trimble Rd. Northbound Northbound Southbound Northbound GROUP 1: LEFT-TURN AND CROSS STREET MOVEMENTS Mainline left-turn change and clearance interval violation...................................................................... 19 5 4 4 6 Mainline left-turn red-light violation ............................ 2 0 0 1 1 Mainline U-turn change and clearance interval violation...................................................................... 9 4 3 2 0 Mainline U-turn red-light violation .............................. 3 0 1 2 0 Cross-street red-light violation ................................... 9 EB WB EB WB EB WB 2 0 1 5 1 0 GROUP 2: RIGHT-OF-WAY AND POSITIONING RELATED Mainline through lane stop bar intrusion ................... 13 9 2 0 2 Mainline left-turn stop bar intrusion ........................... 12 0 2 2 8 Lane change violation ................................................ 1 0 0 1 0 Track intrusion violation ............................................. 2 0 0 2 0 GROUP 3: STOPPING ON TRACKS Vehicle stopped on tracks ......................................... 2 2 0 0 0 Mainline left-turn vehicles queued on tracks ............. 2 2 0 0 0 Cross street left-turn vehicles queued on tracks ....... 28 EB WB EB WB EB WB EB WB 0 28 0 0 0 0 0 0 Cross street through vehicles queued on tracks ....... 6 EB WB EB WB EB WB EB WB 5 1 0 0 0 0 0 0 GROUP 4: PEDESTRIAN OR BICYCLE RELATED Pedestrian standing on tracks ................................... 4 0 1 1 2 Pedestrian intersection violation ................................ 22 7 7 2 6 Pedestrian jay-walking violation ............................... 2 1 1 0 0 Bicyclist mainline left-turn red-light violation .............. 0 0 0 0 0 Bicyclist cross-street red-light violation ..................... 0 0 0 0 0 GROUP 5: LIGHT RAIL VEHICLE RELATED Light rail vehicle violation .......................................... 0 0 0 0 0 Light rail emergency braking ..................................... 0 0 0 0 0 GROUP 6: RISKY BEHAVIOR OBSERVED WITHOUT DIRECT LRT IMPACT Other .......................................................................... 14 65* 6 6 2 Vehicles backing up to change lanes ........................ 17 9 4 4 Right turn or through maneuvers from the left lane ... 64 21 14 29 Left turn from through or right turn lane ..................... 40 9 4 27 Aggressive driver behavior (cutting off, abrupt stops) ........................................................................ 1 0 0 1 Right turn on red that opposes through traffic ........... 1 0 1 0 0 Mainline through red-light violation ............................ 35 NB SB NB SB NB SB 8 2 6 13 5 1 TOTAL (GROUPS 1-6) ............................................ 373 141 74 63 95 * During the data reduction for North First St. and Brokaw Rd. intersection, risky behavior observations under the "RISKY BEHAVIOR OBSERVED W/O DIRECT LRT IMPACT" were not separated into subcategories; however, these observations are included in the total number of observations. A brief summary of the baseline condition data is as follows:

TCRP Web-Only Document 53 30  The most observed risky behavior associated with a light rail maneuver was a left-turn during the yellow change and all-red clearance intervals, and in several cases completing the turn while the red interval was expiring. This behavior occurred consistently at all three intersections.  The type of risky behavior with the highest potential for severe results was entering the intersection on red several seconds after the yellow interval had expired.  Pedestrian risky behavior did not demonstrate a high potential of severe implications. Several pedestrians crossed the intersection either during the Don’t Walk interval, or continued to cross after the Flashing Don’t Walk had expired. Risky behaviors were rare events and tended to be proportional to the frequency of the maneuvers. All three intersections had minimal pedestrian activities.  Risky behavior associated with cross-street motorists entering the intersection on red was minimal; a mere nine violations were observed during the 160 hours of data collection. The westbound movement at Charcot Ave. had the highest frequency of red light violations; five of the total nine observations at all three intersections.  There were no observations of vehicles getting stuck in the track right-of-way.  The intersection of North First St. and Brokaw Rd. had the highest frequency of risky behaviors. Vehicles making westbound left turns onto southbound North First St. often could not clear the intersection due to downstream queuing, which resulted in vehicles stopped on the tracks.  Overall, none of the three intersections demonstrated unusual risky behaviors. ASSESSMENT OF THE OPERATIONAL AND SAFETY ASPECTS OF ALTERNATIVE AND POSSIBLE SUPPLEMENTAL TRAFFIC CONTROL DEVICES The primary means for assessing the operational and safety aspects of alternative and possible supplemental traffic control devices was through the use of focus groups with local drivers. Focus Group Design and Process Focus groups were conducted in June 2009 in San Jose, CA. The focus groups were conducted to explore the opinions of drivers regarding existing traffic control devices as well as a variety of alternative and supplemental traffic signal displays, signs, and pavement markings at intersections where traffic interfaces with light rail transit. Focus groups were conducted in San Jose to obtain input from local drivers with experience driving in proximity to the VTA light rail system. Thirty people (recruited by a local company in the San Jose area) participated in three focus group sessions. Participants were:  Fairly evenly divided by sex – 57 percent were female;  Representative of a range of ages – 60 percent were under 40 years old, 30 percent were between the ages of 40 and 60, and the remaining 10 percent were 60 years old or above;  Long time residents of the San Jose area – almost one-third reported having lived in San Jose their entire lives, and 40 percent lived there over 20 years; and  Well acquainted with the light rail system – nearly 75 percent drove “often” or “frequently” on streets that included light rail and many had used the light rail system (though most do not currently use the system except to attend “crowded” events).

TCRP Web-Only Document 53 31 The focus groups were designed around three different scenarios in which motor vehicles interfaced with LRT at signalized intersections. The three scenarios included:  No Left Turn Allowed Scenario (parallel to the LRT line);  Left Turn Allowed Scenario (parallel to the LRT line); and  Cross-Street Scenario (perpendicular to the LRT line). Participants were shown 27 different simulated scenes from the driver’s perspective in an automobile. The No Left Turn Allowed Scenario is illustrated in Figure 10. In this scenario, participants were told that the driver was traveling parallel to a light rail line and needed to turn left at the first intersection where a left turn was allowed. Under this scenario, participants were shown eight different scenes. In each scene, the left-turn movement was restricted at the intersection, and a passive sign was provided to indicate this restriction to drivers. However, the scenes varied in terms of whether the through movement had circular green or green arrow displays and the type of sign present (three different signs were used, as shown in Figure 10). Figure 10. No Left-Turn Allowed Scenario with Various Combinations of Traffic Signal Displays and Three Different Signs The Left Turn Allowed Scenario is illustrated in Figure 11. In this scenario, participants were also told that the driver was traveling parallel to a light rail line and needed to turn left at the first intersection where a left turn was allowed. Under this scenario, participants were shown ten

TCRP Web-Only Document 53 32 different scenes. In each scene, a left turn was allowed at the intersection, and an active blank- out warning sign was provided to indicate to the driver if a train was approaching. However, the scenes varied in terms of whether the left-turn movement had the green or red arrow signal display, whether the through movement had circular green displays or green arrow displays, whether a train was approaching, and the type of active blank-out sign present (five different activated blank-out signs were used, as shown in Figure 11). Figure 11. Left-Turn Allowed Scenario with Various Combinations of Traffic Signal Displays and Five Different Blank-out Signs The Cross-Street Scenario is illustrated in Figure 12 and Figure 13. In this scenario, participants were told that the driver was traveling on a street running parallel to a light rail line and needed to go through the intersection, crossing over the tracks. Both through movements and left-turn movements were allowed at the intersection shown in this scenario; however, the car was positioned in a through lane. Under this scenario, participants were shown nine different scenes. In each scene, some form of warning sign was present, whether it be passive, active, or both. The scenes varied in terms of whether the left-turn movement had a green or red arrow signal display, whether the through movement had green or red signal displays, whether a train was approaching, and the type of passive and active traffic control devices that were present (two different passive signs were used—a trolley crossing warning sign and the R15-7 regulatory sign,

TCRP Web-Only Document 53 33 as shown in Figure 12—and two different active devices were used—the W10-7 sign (shown in the top picture of Figure 13) and in-pavement lights (shown in the bottom picture of Figure 13). Figure 12. Cross-Street Scenarios with Various Combinations of Traffic Signal Displays and Two Different Passive Signs (Trolley Crossing Warning and R15-7)

TCRP Web-Only Document 53 34 Figure 13. Cross-Street Scenarios with Various Traffic Signal Displays and Two Different Types of Active Traffic Control Devices (W10-7 Sign and In-pavement Lights) Each focus group was divided into the following three activities, shown in the order that they occurred:  Comprehension activity;  Comparison activity; and  Discussion. During the comprehension activity, the participants were shown all 27 scenes. They were told to pay attention to the lane that the car was in and were asked to determine if the driver should stop,

TCRP Web-Only Document 53 35 turn left, or go straight. A response sheet was provided for the participants on which to record their answers for each of the 27 scenes. They were also asked to record, based on the information in each scene, the probability that a train was approaching the intersection. They were asked to record any value between 0 and 100 percent – 100 percent if they were absolutely certain a train was approaching the intersection and 0 percent if they were absolutely certain a train was not approaching the intersection. During the comparison activity, the participants were shown pictures in pairs. Each pair of pictures showed a particular situation with a slight variation in traffic signal displays, signs, or pavement markings. During this activity, participants were asked to decide which of the two scenes was better at conveying to drivers what they should do at the intersection. In each pair of pictures there was a picture labeled “A” and a picture labeled “B.” The participants were provided another response sheet to record their choice. There were a total of 20 comparisons. During the discussion activity, participants were asked specific questions about the traffic control devices they saw during the previous activities. Questions included:  Which of the three traffic signs is the best for indicating to drivers that a left turn is NOT permitted at the intersection?  Which of the five electronic signs is best for indicating to drivers that a train is coming and that a left turn MUST NOT be made at this time?  Which of the two traffic signs is better to indicate to drivers that they are at an intersection that crosses light rail tracks?  Is an electronic train-coming sign helpful in the cross-street situation? Are both the passive traffic sign and the electronic sign needed? Asking these questions during the focus group provided a more in-depth understanding about why participants may not have understood the meaning of particular device and why participants chose one device over another in the comparison activity. Focus Group Results For the No Left Turn Allowed scenario, the results of the comprehension activity are shown in Table 15. Regardless of the type of sign or traffic signal displayed, participants overwhelmingly were aware of the appropriate action, with between 83 and 93 percent responding correctly to “go straight.” Scene 17, which had the R3-2 No Left Turn symbol sign (Figure 10a) and green thru arrow signal displays yielded the highest percentage of correct responses of the eight scenes. Interestingly, when through arrows on the pavement in both lanes were added to this scene, which was done in Scene 23, this yielded one of the lowest percentages of correct responses. Scene 25, the only other scene with through arrow pavement markings, also yielded a mere 83 percent correct responses. With regards to the probability that a train was approaching, it was never known whether or not a train was approaching; therefore, the correct response in all eight scenes was “maybe.” While the majority of respondents gave the correct response in all scenes except Scene 11, fewer participants correctly interpreted that a train might be coming as compared with their understanding of what they should do at the intersection. In examining the findings, there does not seem to be a reasonable explanation as to why the respondents answered the way they did. The traffic signal was always green, which, it was anticipated, would have indicated more to participants that a train would be coming than “definitely not” coming, which was the second

TCRP Web-Only Document 53 36 most common response. One explanation could be that the participants were not familiar with this particular scenario and were confused about the relation between these scenes and the arrival of the trains. It almost appears as though the responses improved as the scenes were shown, indicating that participants may have “learned” to interpret and understand the scenes under this scenario. Table 15. Comprehension: No Left Turn Allowed Scenario Scene Numbers What SHOULD the driver do? Participant’s Response Is there a train coming? Participant’s Response to probability of train approaching Go Stop Definitely Not Maybe Definitely Yes Left Straight 11 Go Straight 0% 93% 7% Not Known 63% 37% 0% 13 Go Straight 10% 83% 7% Not Known 33% 57% 10% 15 Go Straight 3% 93% 3% Not Known 33% 57% 10% 17 Go Straight 0% 97% 3% Not Known 37% 53% 10% 19 Go Straight 10% 87% 3% Not Known 27% 70% 3% 21 Go Straight 7% 87% 7% Not Known 24% 69% 7% 23 Go Straight 3% 83% 14% Not Known 30% 67% 3% 25 Go Straight 3% 83% 14% Not Known 30% 67% 3% The results of the comparison activity are shown in Table 16. The R3-2 No Left Turn symbol sign (Figure 10a) was compared with the version that also depicted railroad tracks (Figure 10b), as well as with the text version of the sign (Figure 10c). The majority of the participants preferred the R3-2 sign in both comparisons. In the discussions, most participants did not like the symbol sign with the tracks because it was not clear what the tracks were; some thought it looked like a ladder. The text version of the sign was the least popular among participants. Some participants commented that not all drivers can read English, and the sign might take longer to process. The R3-2 sign was thought to be simple, with no need for further interpretation or modification. When comparing the traffic signal displays (comparison 4), in general, the green arrow displays were preferred over the circular green displays, and the through arrows on the pavement were generally popular with participants (comparison 5). When the green arrow signal displays were compared with the circular green signal displays and through arrows on the pavement (comparison 6), the majority of participants preferred the circular green signal displays and the arrow pavement markings. When the two different signal displays were paired with through arrows on the pavement (comparison 7), participants were split almost evenly on the arrow versus circular signal displays. In this case, a few participants noted that having arrows on both the signal displays and on the pavement might be unnecessary or “too much.”

TCRP Web-Only Document 53 37 Table 16. Comparisons: No Left Turn Allowed Scenario # Traffic Signal Displays Comparison Traffic Control Devices for Scenario A Traffic Control Devices for Scenario B Which scenario is BETTER at conveying what the driver SHOULD do? A B 1 Green thru signals Passive Signs R3-2 sign No Left Turn Across Tracks (symbol) 53% 47% 2 Green thru signals Passive Signs R3-2 sign No Left Turn Across Tracks (text) 73% 27% 3 Green thru signals Passive Signs No Left Turn Across Tracks (text) No Left Turn Across Tracks (symbol) 70% 30% 4 Green thru signals Thru signals Circular greens Green arrows 37% 63% 5 Green thru signals Pavement markings No pavement markings Through arrows on pavement 3% 97% 6 Green thru signals Thru signals / pavement markings Green arrows/ no pavement markings Circular greens / through arrows on pavement 30% 70% 7 Green thru signals Thru signals Green arrows / through arrows on pavement Circular greens / through arrows on pavement 53% 47% For the Left Turn Allowed scenario, the results of the comprehension activity are shown in Table 17. Overall, an overwhelming number of participants knew what they should do in each of the 27 scenes. However, the blank-out signs shown in scenes 20, 22, and 24 (Figure 11c, d, and e, respectively) were not well understood by participants as compared to the W10-7 sign (shown in scenes 10, 16, 18, 26, and 27). While a majority of participants knew they should stop (there was a red left-turn arrow in each of these three scenes), less than a third of the participants knew for certain that a train was approaching when these signs were displayed. For the most part, it would appear that most participants did not realize these signs were meant to replace the W10-7 signs. This result may be because these three signs are unfamiliar to drivers in the San Jose area, who are accustomed to the W10-7 train-coming sign.

TCRP Web-Only Document 53 38 Table 17. Comprehension: Left Turn Allowed Scenario Scene Numbers What SHOULD the driver do? Participant’s Response Is there a train coming? Participant’s Response to probability of train approaching Go Stop Definitely Not Maybe Definitely Yes Left Straight 10 Stop 0% 0% 100% Yes 0% 6% 93% 12 Turn Left 100% 0% 0% No 87% 13% 0% 14 Turn Left 100% 0% 0% No 77% 23% 0% 16 Stop 0% 3% 97% Yes 0% 3% 97% 18 Stop 0% 0% 100% Yes 3% 0% 97% 20 Stop 0% 3% 97% Yes 13% 63% 23% 22 Stop 0% 10% 89% Yes 13% 53% 33% 24 Stop 7% 13% 80% Yes 7% 63% 30% 26 Stop 0% 3% 97% Yes 0% 3% 97% 27 Stop 0% 3% 97% Yes 0% 0% 100% The other discrepancies in the responses for probability of a train approaching are not as clear. It is possible that a few people were aware that a train could technically be approaching even if the left turn movement had the green (in San Jose, the intersections are not pre-empted by the train; therefore, sometime the train must stop at the intersection). As such, while the left turn green arrow shown in scenes 12 and 14 was meant to indicate to drivers that a train was not approaching, it is not actually incorrect to respond that a train might be approaching, as did some participants. A few participants reported that they do not trust the lights. One participant said that even though there is a green light, he always assumes there is a train coming, “just to be cautious.” The results of the comparison activity for the Left Turn Allowed scenario are shown in Table 18. With regard to the W10-7 sign, the addition of the word “Train” seemed to be an improvement over the icon alone (80 percent of participants preferred the sign with the word “Train” in comparison number 10). There was a comment in the discussion about how the sign with the word “Train” makes it larger and more impressive, so there is a possibility that the size, not the message, may have swayed some of the participants. When comparing the W10-7 with the word “Train” to the VTA’s new active train-coming sign, which alternates between the train icon and the No Left Turn symbol (comparison number 14), 63 percent of participants preferred the alternating sign. This is an interesting result when considering that the No Left Turn symbol activated blank-out sign, on its own, was not very popular when compared to the W10-7 (comparison number 11). One participant pointed out that if the alternating sign included the word “Train” it would be even better. Another participant noted that, because of the No Left Turn symbol, this sign could be confusing about whether you could ever make a left turn at this intersection. Overall, however, it appears from the results that the majority of participants favored the VTA’s new alternating train-coming sign over all other options.

TCRP Web-Only Document 53 39 Table 18. Comparisons: Left Turn Allowed Scenario # Traffic Signal Displays Comparison Traffic Control Devices for Scenario A Traffic Control Devices for Scenario B Which scene is BETTER at conveying what the driver SHOULD do? A B 8 Green thru and left signals Thru signals Circular greens Green arrows 60% 40% 9 Green thru signals with Red turn signal Thru signals Circular greens Green arrows 60% 40% 10 Green thru signals with Red turn signal Active blank- out signs W10-7 W10-7 with word “Train” 20% 80% 11 Green thru signals with Red turn signal Active blank- out signs W10-7 No Left Turn (symbol) activated blank-out sign 77% 23% 12 Green thru signals with Red turn signal Active blank- out signs No Left Turn Across Tracks (symbol) activated blank-out sign No Left Turn (symbol) activated blank-out sign 50% 50% 13 Green thru signals with Red turn signal Active blank- out signs No Left Turn Across Tracks (symbol) activated blank-out sign R3-2a activated blank- out sign 70% 30% 14 Green thru signals with Red turn signal Active blank- out signs W10-7 with word “Train” W10-7 alternating with No Left Turn (symbol) activated blank-out sign 37% 63% 20 Green thru signals with Red turn signal Existing versus improved conditions (per VTA project) W10-7 with “Train” W10-7 alternating with No Left Turn (symbol) activated blank-out, “Keep Clear” and bull nose pavement markings, and bollards 23% 77% The improvements currently in progress by the VTA at several intersections across their system were illustrated in scene number 27, and participants were asked to compare the existing conditions to these improved conditions (comparison number 20). The participants comprehended the situation almost perfectly (as discussed above), and the majority (77 percent) thought the improvements would better convey to drivers what to do. In the discussion section, everyone seemed to prefer the improvements over the existing conditions. According to the comments, the yellow bull nose pavement markings and the “Keep Clear” pavement marking were big improvements over the existing conditions. The general consensus was this improved situation would make most people stop farther back from the intersection at the stop bar. Another interesting comparison under this scenario was that the results of the circular green signal displays versus the green arrow signal displays (comparisons 8 and 9). The responses were split 60/40, with a slight majority preferring the circular green displays in both comparisons. Some reasons given for this result were that the circular green displays stand out more than the green arrow displays (they cover more surface area) and that too many arrows made the situation too overwhelming and confusing. The participants that liked the arrows reported that they provide drivers with more information and reinforced to drivers what they were supposed to do.

TCRP Web-Only Document 53 40 For the Cross-Street Scenario, the results of the comprehension activity are shown in Table 19. In all of the eight scenes, the through signals were red and the correct driver action was to stop; however, different left-turn displays and various active warning devices were used to indicate to participants whether or not a train was approaching the intersection. In only two scenes (scenes 2 and 4) was the approach of a train unknown based on the traffic control devices. As with the other scenarios, a large majority of participants understood what they should do at the intersection. There did seem to be some confusion for a small percentage of the participants about whether the correct action was to stop or go (straight or left); however, this was likely a function of the design of the focus group activity not the traffic control devices. With the Cross- Street scenario, participants did appear to be less aware or confident of the meaning of the traffic control devices in relation to whether or not a train was approaching than in the scenarios on the parallel street. This finding could be a function of the predictive priority system used in San Jose (sometimes the train does have to stop at the intersections), as discussed previously. It does appear that some type of active device on the cross-street could help clear up some of the uncertainty drivers appeared to have on the cross-street as to the arrival of a train. The results of the comparison activity are shown in Table 20. The trolley crossing warning sign used in San Jose was compared to the MUTCD R15-7 divided highway regulatory sign (comparisons 15 and 18). The majority of participants preferred the trolley crossing over the R15-7 sign. The participants preferred the warning sign because it was yellow and would attract the driver’s attention. They also thought the R15-7 sign was confusing and did not necessarily give the message that there were in tracks and not just a wide median. However, many participants felt the trolley crossing warning sign could be improved by saying instead “Train Crossing.” This change would make the sign seem more serious and less “friendly looking.” Table 19. Comprehension: Cross-Street Scenario Scene Numbers What SHOULD the driver do? Participant’s Response Is there a train coming? Participant’s Response to probability of train approaching Go Stop Definitely Not Maybe Definitely Yes Left Straight 1 Stop 7% 10% 83% No 63% 36% 0% 2 Stop 3% 3% 93% Not Known 20% 64% 17% 3 Stop 10% 7% 83% No 57% 42% 0% 4 Stop 3% 3% 93% Not Known 23% 67% 10% 5 Stop 13% 3% 83% No 53% 40% 7% 6 Stop 0% 3% 97% Yes 7% 10% 83% 7 Stop 10% 7% 80% No 67% 29% 3% 8 Stop 0% 0% 100% Yes 7% 3% 90% 9 Stop 0% 3% 97% Yes 7% 64% 30%

TCRP Web-Only Document 53 41 Table 20. Comparisons of Different Intersection Treatments: Cross-Streets # Situation Comparison Traffic Control Devices for Scenario A Traffic Control Devices for Scenario B Which scene is BETTER at conveying what the driver SHOULD do? A B 15 Red thru signals with Green turn signal Passive signs Trolley crossing warning sign R15-7 divided highway sign 63% 37% 16 Red thru signals with Green turn signal With or without W10-7 sign (with trolley crossing sign) Without W10-7 sign With W10-7 sign 10% 90% 17 Red thru and left signals R15-7 divided highway sign with or without W10-7 sign Without W10-7 sign With W10-7 sign 7% 93% 18 Red thru and left signals Passive signs (with active warning device) Trolley crossing warning sign R15-7 divided highway sign 70% 30% 19 Red thru and left signals Different active warnings (with trolley crossing sign) W10-7 sign In-pavement lights 83% 17% When comparing the existing conditions without an active warning device on the cross-street to a scene with a W10-7 sign mounted on the mast arm (comparisons 16 and 17), the overwhelming majority of participants preferred the scene with the W10-7 sign. All the participants generally agreed that the active blank-out signs were helpful on the cross-street. Most people agreed that the extra information was useful and would better catch the driver’s attention, although some felt that it was not necessary because the light was already red. A couple of participants noted that the active signs could also be helpful for pedestrians as it would stop them from crossing the main street when a train is coming. In comparison 19, the W10-7 sign was compared with in-pavement lights (similar to those used in Houston). A few of the participants (17 percent) indicated they preferred the in-pavement lights over the W10-7 sign. Most did not understand the in-pavement lights (several did not even notice them), and they felt that there would need to be a supporting sign explaining the meaning of the lights. This result could have been partially a function of the way the lights looked in the simulated scene. SUMMARY AND CONCLUSIONS REGARDING OPERATIONAL AND SAFETY ASPECTS OF EXISTING, ALTERNATIVE, AND POSSIBLE SUPPLEMENTAL TRAFFIC CONTROL DEVICES The operational and safety aspects of existing, alternative, and possible supplemental traffic control devices associated with controlling traffic at signalized intersections that interface with light rail transit were assessed as part of this research project. Existing traffic control devices at the three study intersections consist of conventional traffic signals, W10-7 train-coming activated blank-out signs for the left turn movements on North First St. (mounted adjacent to the left-turn signals), “RR Xing” pavement markings in the left-turn pockets on North First St., and “Trolley Xing” warning signs on the cross-street approaches. To assess the operational and safety aspects of these existing traffic control devices, the research team gathered and analyzed data that would indicate how well the intersections currently operate and the existing level of safety at the intersections. These data consisted of historical crash and

TCRP Web-Only Document 53 42 near-miss incident data and observations of operations specifically, driver behaviors at the intersections. The team also assessed the operational and safety aspects of alternative and possible supplemental traffic control devices that might be used at the intersections to improve driver situational awareness and safety in general and under increased LRV speed conditions. Alternative and possible supplemental traffic control devices assessed included: a variety of alternative (and possible supplemental) train-coming activated blank-out signs, a variety of passive warning and regulatory signs, and alternative pavement markings. To assess these traffic control devices, the team conducted a series of focus groups with local San Jose drivers to obtain their feedback on the devices. With regard to the existing traffic control devices, both the historical data and the observations of driver behaviors made at the three intersections show that the intersections currently operate well, with minor safety concerns. Drivers appear to understand the operation of the intersection with the LRT and generally respect and comply with the traffic control devices at the intersections. The primary crash type tends to be read-end crashes, followed by right-angle crashes. Rear-end crashes generally have little to do with traffic control device compliance (they are usually a result of speeding and drivers not paying attention) and they have almost no impact on LRT operations or safety. Right-angle crashes, on the other hand, can be indicative of problems with device compliance (e.g., drivers running red lights) and can have direct impact on the occurrence of LRT-motor vehicle crashes (when the red-light running is occurring on the cross-streets). However, right-angle crashes make up a small portion of the overall crashes at the study intersections. Of the 109 crashes that occurred at the intersections in the past 3 years, there were 19 right-angle crashes (17 percent). In addition, the data show that most of the red-light running is occurring on North First St., not the cross-streets. In fact, there were three crashes in the past 3 years (one at each intersection) where a cross-street driver was cited for running a red light). This is also supported in the near-miss incident and observational data. In a week’s worth of observational data (160 hours), there were nine observed red-light violations on the cross- streets (approximately three per intersection) as compared with 35 observed on the main street (North First St.) through red-light violations. The primary operational and safety concern with regard to the traffic control devices appears to be the left-turn movements from North First St. onto the cross-streets; however the concerns are relatively minor. According to the crash data provided by the VTA, there were 11 LRV-motor vehicle crashes at the 3 intersections over the past 3 years (approximately 2 crashes per year at the intersection of Brokaw Rd., 1 crash per year at the intersection of Trimble Rd. and less than 1 crash per year at the intersection of Charcot Ave). Ten of the 11 crashes were a result of a left- turn violation. The driver observational data show that there were only five left-turn drivers observed running a red-light in the 160 hours’ worth of data. It is not clear whether the North First St. left-turn violations are a question of driver misunderstanding or confusion, or one of blatant noncompliance. The low number of violations and crashes suggests that it is not the former. If drivers were confused about what to do, the violation and crash rates would likely be much higher. This interpretation is supported by the results and feedback received in the focus groups. While some participants did admit that North First St. can be confusing (especially for drivers not familiar with the situation), none of the participants was confused about what to do in the three left-turn allowed scenes with the existing traffic control devices (i.e., 100 percent comprehension

TCRP Web-Only Document 53 43 about what to do). In addition, the feedback provided was that the existing traffic control devices are clear at conveying to drivers what they are supposed to do. With regard to the alternative and possible supplemental traffic control devices, a number of findings stand out. First are the findings from the comparisons of the existing W10-7 sign to alternative train coming activated blank-out signs. Participants did seem to prefer the version of this sign with the word “Train” to the existing W10-7 sign (the sign with “Train” was preferred by four times as many participants, although there was no consensus on this). The fact that drivers familiar with LRT operations preferred the sign with the word “Train” could indicate that it could result in an operational and/or safety improvement over the existing sign. Probably the most significant finding of the focus groups was the result and feedback received regarding the overall VTA improvements being made at all three study intersections. When the W10-7 sign with the word “Train” was compared to VTA’s new alternating blank-out sign, twice as many participants preferred VTA’s new sign, suggesting that it could be even a bigger improvement over the current sign. When comparing the overall VTA improvements (including pavement markings, recessed stop bar, and bollards) to the existing conditions (including the word “Train” on the W10-7), participants preferred the VTA improvements by nearly three to one. It appears that, based on the results of the focus groups, the VTA improvements are expected to result in improved operational and safety conditions. With regard to the W10-7 sign on the cross-street, most participants thought it was helpful, but not necessary. Participants mentioned that at busy intersections it could be an improvement because you cannot always see the passive sign (it might be blocked by other cars or be mounted on one side of the road making it hard for drivers to see in the far lanes). Also, in cases when the signal phase is longer than normal, drivers may wonder why and start to get impatient. The W10-7 sign on the cross-street might inform them it is taking longer than normal because a train is coming. Recommended Traffic Control Devices Based on analyses of the historical safety data, observations of driver behaviors, and focus group findings, the supplemental and alternative traffic control devices being implemented by the VTA appear to address the issues of most concern at the study intersections (i.e., left-turn violations on North First St.). The VTA’s enhancements focus on changing driver behaviors during left-turn movements from both North First St. and the cross-streets that can lead to left-turn collisions and track intrusions, respectively. While the occurrence of these incidents was relatively low, the enhancements being made to the intersections should help further mitigate against these types of incidents in the future. The VTA’s improvements do not include traffic control devices aimed at cross-street traffic. That is because the data (both historical safety data and observational driver behavior data) do not necessarily warrant such treatments. As stated previously, right-angle crashes represented a mere 17 percent of the crashes at these intersections over the past 3 years, and most of the red- light running appears to be occurring on North First St. as opposed to the cross-streets. Likewise, in the observational data, there were nine observed cross-street red-light violations as compared with 35 observed North First St. through red-light violations. In addition, focus group participants did not feel that the installation of the W10-7 signs on the cross-street was a necessary safety measure in San Jose.

TCRP Web-Only Document 53 44 However, while the crash and observational data may not necessarily warrant treatment on the cross-streets in San Jose, the research team would recommend that in some locations, the installation of W10-7 signs on the cross-street approaches would be a prudent countermeasure if train speeds were to be increased to 40 mph. The primary reason for this recommendation is to provide an active warning to drivers on the cross-street regarding the arrival of an LRV. Currently, when trains are operating in excess of 35 mph, the MUTCD requires the use of flashing lights on the cross-street and recommends the use of crossing gates. Therefore, the W10-7 signs could be used in lieu of flashing lights to provide the active warning to the cross- street drivers. Secondly, focus group participants were supportive of providing the W10-7 signs on the cross-street, especially at wide, busy intersections. In both comparisons of a cross-street scene without a W10-7 sign to one with a W10-7 sign, 90 percent or more of participants chose the scene with the W10-7 sign as the better option for conveying to drivers what they should do at the intersection. As the use of W10-7 signs for left-turning vehicles has been effective at mitigating left-turn collisions, some agencies (e.g., TriMet) have also found them to be effective at mitigating right-angle collisions due to red-light running on the cross-street approaches. Another concern associated with the increase in LRV speeds above 35 mph is pedestrians. As VTA’s improvements aim to mitigate against left-turn crashes and track intrusions, they are not focused on pedestrians. Similar to the cross-streets in San Jose, data that were reviewed in relation to pedestrian operations and safety in San Jose do not necessary warrant special treatments for pedestrians. Pedestrian volumes at the three study intersections are extremely low, and there has not been an LRV-pedestrian crash at any of the three study intersections in the past 3 years. However, in consideration of other areas that may experience higher pedestrian volumes or higher LRV-pedestrian crash rates, the research team would recommend the consideration of pedestrian countdown signals at crossings where LRV speeds are increased above 35 mph. This recommendation is of particular importance at wide intersections to give pedestrians information on how much time they have to cross. This information can help pedestrians make better decisions about whether or not to start crossing and could help reduce the number of pedestrians trapped in the middle of the road. As an added note, several focus group participants indicated that installing W10-7 signs on the cross-street (within view of the pedestrians) may also help alert pedestrians when a train is coming. Preliminary Cost Estimates of Traffic Control Devices Cost estimates were obtained from the VTA for the Light Rail Left-Hand Turn and Track Intrusion Project – Phase I, which includes improvements to nine intersections, with a very similar type of work at each intersection. The cost for all improvements was approximately $360,000, an average of about $40,000 per intersection. Therefore, for the three intersections in this study, the total cost was approximately $120,000. The current edition of the MUTCD requires the use of flashing lights and recommends the use of flashing lights at LRT crossings where LRVs operate at speeds higher than 35 mph. As part of Task 3, the TCRP A-32 research team was to compare the costs of traffic control devices or safety countermeasures that could be used in lieu of flashing lights and crossing gates to allow LRVs to operate at speeds above 35 mph. A typical gated LRT crossing includes four gates at the intersection and flashing lights on each approach, one gate on each of the side streets, and one gate for each left-turn movement on the

TCRP Web-Only Document 53 45 major street. All gates require signal communications and gate assembly equipment. Typical gate assemblies are manufactured to accommodate gates up to 40 feet long. Cost information obtained from the Maryland Department of Transportation and the Maryland Transit Administration for gate assembly, flashing lights, controller and cabinet, communications, and track circuit detection revealed that the cost for one intersection would be in the range of $550,000. With regard to the additional recommendation (in some locations) for W10-7 signs on the cross- street approaches, the cost to install two W10-7 signs at an intersection would be approximately $7,000, assuming that they can be installed on the existing mast arms. With regard to the recommendation (in some locations) for the installation of pedestrian countdown signals to replace the existing conventional pedestrian signal heads, the cost to retrofit each intersection is approximately $10,000 for eight signal heads. Table 21 shows a summary of these costs. Table 21. Cost Estimate Summary for Three Intersections Item Cost Per Intersection Left-turn and track intrusion enhancements (per VTA) $40,000 Flashing light / signal assembly $125,000 Gate assembly $425,000 W10-7 signs on cross-street approaches, 2 signs $7,000 Retrofitting pedestrian head with countdown signals, 8 signal heads $10,000 Based on the estimated costs presented above, the cost ratio of the flashing lights and gates scenario to the VTA improvements is nearly 14 to 1. Even with the VTA improvements, the W10-7 signs on the cross-street, and pedestrian countdown signals on all four approaches, the cost ratio of the flashing lights and gates scenario to these improvements is still nearly 10 to 1. Finally, it is important to note that sign modifications in California, are subject to additional requirements established by local regulatory agencies (California Public Utilities Commission) that would also have to agree to modify their traffic control device standards in order for the higher speeds to be implemented.

Next: Chapter 4: Modeling/Simulation of Light Rail Operating Conditions: North First Street from East Brokaw Road to Trimble Road »
Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH Get This Book
×
 Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s Transit Cooperative Research Program (TCRP) Web-Only Document 53: Operation of Light Rail Transit through Ungated Crossings at Speeds over 35 MPH presents the findings of a micro-simulation modeling study that explored the impacts of higher light rail vehicle speeds on intersection safety.

READ FREE ONLINE

  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!