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.
PeMS has a built-in diagnostic feature that identifies the percentage of sensor data that is âgood.â Good data are those that have been successfully transmitted from sensors in the field into the PeMS system and that meet other diagnostic criteria (e.g., speeds not excessive and reported flows within acceptable traffic engineering ranges). The circles in Table 3.1 indicate the relative availability of data, with green circles corresponding to facilities that have roughly 70 percent or more good data available, while red circles represent facilities with less than 50 percent of data available. Blue circles lie between these two thresholds. The next-to-last column in Table 3.1 provides an indication of current travel time reliability along the facility. The study team wanted to select facilities with at least a reasonable level of unreliability, or there would be no reliability problems for investments to solve or reliability benefits for the SHRP 2 analysis tools to measure. This assessment was compiled by looking at 2012 weekday travel time reliability along the facility. Travel time reliability was measured using the travel time index (TTI) as calculated in PeMS as well as calculated externally in a database using a âwalk the time-space matrixâ method. This method and the teamâs calculations are described in Section 4.4 The analysis shown in Table 3.1 allowed the study team to narrow the potential test facilities from nine facilities to four based on a review of detection quality, reliability levels, and potential facility solutions. The team initially chose Interstate 210 (I-210) in Los Angeles County and State Route 57 (SR-57) in Orange County. I-405 was selected as an alternate pilot site in either county. The team wanted to select a facility from each county to provide a wider range of travel conditions and stakeholders. Further consultation with Caltrans allowed the study team to select two final facilities for the pilot testing: I-210 in Los Angeles County and I-5 in Orange County. The I-210 facility has the worst reliability among the facilities considered. It also has a mixture of urban and suburban development. The I-5 facility replaced the SR-57 facility as the facility selected in Orange County. Although SR-57 had a nice combination of data quality and poor reliability, the majority of improvements along the facility were already programmed, so analysis of the facility would not help future funding decisions. The I-5 facility offers right-of-way for potential expansion projects as well as relative flexibility in funding future projects. The team also considered including I-10 and I-110 in Los Angeles County, but these facilities were rejected due to the potential conflict with ongoing express lane demonstrations. 3.2 Final Test Facilities The next two sections provide background information from the CSMPs for the two facilities selected as the focus for the SHRP 2 reliability pilot testing. I-5 Facility in Orange County The first facility selected is the I-5 facility in Orange County. As shown in Figure 3.2, I-5 is a 45- mile facility with four to six general-purpose (or mainline) lanes in each direction. There are also 27
one to two high-occupancy vehicle lanes in each direction. As is the standard configuration in Southern California, these high-occupancy vehicle lanes are barrier-separated, so access is limited to designated locations along the freeway. As described in Chapter 7, this led to significant difficulty in calibrating a FREEVAL-RL model for the facility. Figure 3.2. Map of I-5 facility in Orange County. 28
There are auxiliary lanes throughout the I-5 facility to improve merging to and from closely spaced ramps and connectors. Some short segments have up to three auxiliary lanes in a single direction. The I-5 facility has a very complex geometry that proved to be difficult to model in the SHRP 2 tools. Figures 3.3 and 3.4 provide speed contour plots from the I-5 CSMP (Caltrans 2012). As shown in the figures, congestion is heaviest in the northbound direction during the p.m. peak period followed by congestion in the a.m. peak period for the southbound direction. During the most congested part of the day, congestion can extend 12 to 15 miles with queuing from downstream bottlenecks overtaking upstream bottleneck locations. Congestion during an individual peak can last 4 to 6 hours with some portion of the facility congested nearly the entire day. 29
Figure 3.3. Speed contour plots for northbound I-5 in November 2010. Source: Caltrans, I-5 CSMP. 30
Figure 3.4. Speed contour plots for southbound I-5 in November 2010. Source: Caltrans, I-5 CSMP. 31
This complex pattern of queuing and congestion is due to 19 major and minor bottlenecks along the facility, which are shown in Figure 3.5. In some cases, the bottlenecks are hidden, so they appear only under certain travel conditions. The I-5 CSMP contains an extensive discussion of each bottleneck and the causes of its formation. Figure 3.5. Map of bottleneck areas on I-5 facility in Orange County. Source: Caltrans, I-5 CSMP. Figures 3.6 and 3.7 show the travel time variation that occurs on the I-5 facility by time of day. The northbound direction has the highest average travel time during the 5:00 p.m. hour (60 minutes on average) as well as the highest travel time variability (approximately 78 minutes). There is also significant unreliability during the a.m. and midday periods in the northbound direction. The northbound and southbound directions do not mirror one another. The southbound direction has a.m. and p.m. peaking, but with average travel time and variability. The I-5 CSMP identifies a number of potential improvements to address the facility bottlenecks (Caltrans 2012). 32
These strategies were tested in the CSMP using microsimulation modeling to understand the effect on mobility, but reliability impacts were not modeled. Figure 3.6. Travel time variation on northbound I-5 in 2010. Source: Caltrans, I-5 CSMP. 33
Figure 3.7. Travel time variation on southbound I-5 in 2010. Source: Caltrans, I-5 CSMP. Figure 3.8 summarizes the scenarios tested in the microsimulation modeling for the I-5 CSMP. Each scenario represents a bundle of improvements that builds on the improvements made in the previous scenario. For example, Scenarios 1 and 2 include a number of interchange, ramp, and auxiliary lane improvements to improve baseline conditions. Scenarios 3 and 4 build on these investments by adding adaptive metering strategies. 34
Figure 3.8. Scenarios tested in CSMP for I-5 facility. Source: Caltrans, I-5 CSMP. In the microsimulation modeling, the scenarios were analyzed for short-term impacts on a 2010 base year and the longer-term impacts on a 2020 horizon year. Scenario 7 was modeled for the horizon year only because the combination of mainline and high-occupancy vehicle additions could not be constructed in the short term. In addition, the CSMP modeling included testing of an enhanced incident management strategy. The SHRP 2 pilot testing examined the reliability impacts of these scenarios. I-210 Facility in Los Angeles County The second facility selected for pilot testing is a 16-mile congested urban segment of I-210 in Los Angeles County. Figure 3.9 shows the entire 45-mile facility covered by the I-210 CSMP. However, the CSMP notes that PeMS detection is available for only the 20 miles in the congested urban area east of I-110 (SCAG and Caltrans 2010). The travel conditions described in the CSMP and in this section are for those 20 miles. The study team decided to focus on the testing of the SHRP 2 reliability products on a slightly smaller 16-mile segment that extends from I-110 to just east of I-605. As shown in Figure 3.9, the 20-mile congested urban area has four to five general- purpose lanes as well as barrier-separated high-occupancy vehicle and auxiliary lanes in each 35
direction. The study team divided its reliability testing into the five-lane section from I-110 to SR-19 (Rosemead Boulevard) and the four-lane section from SR-19 to South Azusa Avenue. Figure 3.9. Map of I-210 facility in Los Angeles County. Source: SCAG and Caltrans, I-210 CSMP. Figures 3.10 and 3.11 show speed contour plots for the 20-mile congested urban area from the I-210 CSMP (SCAG and Caltrans 2010). As shown in the figures, the longest peak period is roughly six hours for both directions of the freeway. In the eastbound direction, traffic is heaviest during the p.m. peak period. In the westbound direction, traffic is heaviest during the a.m. peak period, with some congestion occurring during the p.m. peak period. Congestion is highly directional along the facility due to the location of major job centers in the western portion of the facility and residential communities in the eastern portion of the facility. The study team focused its tool testing on the eastbound direction to capture travelers returning home from work. 36
Figure 3.10. Speed contour plots for eastbound I-210 in April and November 2006. Source: SCAG and Caltrans, I-210 CSMP. 37
Figure 3.11. Speed contour plots for westbound I-210 in April and November 2006. Source: SCAG and Caltrans, I-210 CSMP. 38
The I-210 CSMP identified 17 bottlenecks within the 20-mile urban congested area (SCAG and Caltrans 2010). Figure 3.12 shows the location of bottleneck areas during the p.m. peak period. The a.m. bottlenecks are a subset of these locations. The I-210 CSMP provides a description of the bottlenecks and a detailed causality analysis. Figure 3.12. Map of p.m. bottleneck areas on the I-210 facility in Los Angeles County. Source: SCAG and Caltrans, I-210 CSMP. Figures 3.13 and 3.14 show the travel time variation that occurs along the I-210 facility within the congested urban area. The figures reflect the highly directional traffic flows that occur on the facility. Congestion and unreliable travel occurs during the morning as travelers commute to work in the westbound direction and during the afternoon as people return home in eastbound direction. 39
Figure 3.13. Travel time variation on eastbound I-210 in 2009. Source: SCAG and Caltrans, I-210 CSMP. Figure 3.14. Travel time variation on westbound I-210 in 2009. Source: SCAG and Caltrans, I-210 CSMP. 40
Figure 3.15 shows the many strategies tested in the microsimulation modeling for the I- 210 CSMP. As with the I-5 facility, the strategies were tested for base and horizon years. Unlike the I-5 facility, a 2006 base year was used for the I-210 CSMP due to data availability. The study team decided to use a 2010 base year for consistency with the I-5 facility. The next chapter describes the data compilation for the two facilities in more detail. Figure 3.15. Scenarios tested in CSMP for I-210 facility. Source: SCAG and Caltrans, I-210 CSMP. 41
