Managing traffic through Signal Performance Measures in Pima County

Transcription

1 CASE STUDY Miovision TrafficLink Managing traffic through Signal Performance Measures in Pima County TrafficLink ATSPM Case Study

2 Contents Project overview (executive summary) 2 Project objective 2 Overall improvement 2 Miovision TrafficLink hardware overview 3 Signal Performance Measures, before, during, and after 7 Travel time evaluation 11 Progression quality 12 Conclusion 13 1

3 Project overview Executive summary Pima County was facing a construction project along a busy corridor that would take up to 24 months to complete. The Department of Transportation s (DOT) Traffic Engineering team found that their traditional approach of collecting the necessary traffic data to understand the impact of the construction changes, combined with the process of adjusting signal timing was slow and often incomplete. A lack of data, and inefficient signal timing were leading to poor vehicle progression along the corridor, which increased travel times for citizens. To expedite their maintenance and operations concerns, Pima County implemented a permanent solution that s able to monitor, analyze, and solve traffic issues as construction projects progress along the corridor. Miovision TrafficLink was installed at four key intersections, immediately providing the DOT with rich traffic data insights, which enabled them to improve mobility and normalize travel time along the corridor for the duration of the construction project. Project objective Evaluate and analyze the before, during, and after effects of traffic conditions/signal timing changes of the W Ina Rd. corridor intersections through Signal Performance Measures (SPMs). Overall improvement Increase in cycle length and prioritizing the NS corridor improved the performance on both the EW and NS approaches Minimal effect on travel time on the W Ina Rd. corridor, both median and variability Progression has helped increase Arrivals on Green for the NS approaches Mid-day free mode operation has helped decrease the split failures compared to the 100s constant cycle in the previous period Miovision TrafficLink hardware overview Miovision provided Pima County with SmartLink and Smartview 360 hardware devices, and TrafficLink software to monitor the corridor, and provide Automated Traffic Signal Performance Measures (ATSPMs), and further detection capabilities. Miovision SmartView 360 Miovision SmartLink Miovision SmartSense 2

4 Pilot summary Pima County installed Miovision TrafficLink along the corridor of W Ina Rd. (Figure 1). Based on Signal Performance Measures provided, and estimations with regards to the shift in driving patterns, Pima County has implemented new signal timing plans to accommodate these changes. This report summarizes the performance effects of the freeway interchange closure and signal timing changes on the W Ina Rd. corridor using Miovision TrafficLink. Figure 1. Map of intersections with Miovision TrafficLink installations Comparison of Origin Destination patterns Miovision TrafficLink provides the capability to illustrate Origin Destination (OD) behavior using the MAC addresses captured at each equipped intersection. A comparison of OD patterns before and after the freeway access closure, shown in Figures 3 and 4, illustrate changes in route choice behavior due to this change. The circles on the map at each intersection show the percentage of vehicles last seen (destination) in the corridor originating from the N La Cañada Dr./W Ina Rd. intersection located on the far right end of the study corridor. A more comprehensive deployment (on the north and south intersections) would have provided more insights into the traffic pattern changes on the entire road network. As visible before the closure, a majority of the vehicles traveled the entire corridor (most likely to enter the freeway) while after the closure, corridor length trips are reduced. Figure 3. Feb 1 Origin-Destination patterns on W Ina Rd. 3

5 Figure 4. April 5 Origin Destination patterns on W Ina Rd. In order to reduce bias in the analysis and allow sufficient time for traffic to adapt to the changes implemented, at least one week post-interchange closure is omitted from the analysis. The intersection performance measures provided by TrafficLink are used to compare the effects of the interchange closure prior to its occurrence, during the closure (signal timings were not adjusted), and after the signal retimings. Green represents improvement, red indicates decline in performance, and gray shows no change or minimal change in the performance in terms of split failures. This comparison is shown in Figure 5 and 6. At the intersection of W Ina Rd./La Cañada Dr. the AM peak EB through traffic conditions improved compared to before the pre-closure (no signal changes) period. In Figure 5, we see a summary of the signal performance changes in terms of split failures, for the time period after the interchange closure under normal conditions. Figure 6 compares the post-retiming period without retiming in place, including the post-interchange closure. These charts serve as a comparison between these three time periods and do not provide information/judgement on the actual performance of the intersections during these time periods. The individual intersection performances and changes are explained in more detail further in the report. Figure 5. Freeway interchange closed no signal change vs normal conditions Figure 6. Freeway interchange closed, signal changed vs freeway interchange closed, no signal change 4

6 Using the volume graphs provided in TrafficLink, the changes in the total daily traffic volume before and after the interchange closure are summarized in Table 1. As expected, the volumes on the EW approaches have decreased, and the volumes on the NS approaches have increased. The two far-left intersections, N Shannon and N Camino de la Tierra, overall have experienced larger volume changes compared to La Cañada Dr. and La Cholla Rd. Table 1. Total daily volume changes before and after the interchange closure Intersection of W Ina Rd. and N La Cañada Dr. A one-week comparison of split failures using the TrafficLink Split Trend graphs shows that the interchange closure (prior to the retiming) improved the performance of the EB and WB approaches in terms of the percentage of split failures. The changes in EB and WB volumes, displayed in Table 1, further validates this observation. After the signal timing changes were made, a split trend analysis from TrafficLink shows that the performance of the EB left turn and WB left turn movements during the AM peak, and the SB left turn and NB left turn movements during the PM peak declined. It s important to mention that the signal timing changes improved the NB through and SB through traffic conditions which were also an issue prior to the interchange closure. As shown in Figure 7 for the NB through movement, the TrafficLink Split Trend graphs clearly reveal that the NBthrough and SB-through movements were already experiencing performance issues (i.e., high split failure percentages) even before the closure/timing changes. It is understood that due to the interchange closure, the NB and SB approaches were the main movements expected to provide access to Hwy 10, and were therefore prioritized. However, a comparison of the Split Trend graphs and traffic volumes before and after the closure (but before timing changes), shows that the interchange closure only slightly increased the volume and the number of split failures on the NB and SB approaches at this intersection. The change in cycle length and prioritization of the NB and SB approaches have reduced the effects of a problem that already existed prior to the interchange closure. Figure 7. La Cañada Dr. NB through performance comparison (before) 5

7 Figure 8. La Cañada Dr. NB through performance comparison (during) Figure 9. La Cañada Dr. NB through performance comparison (after) As illustrated in Figure 10, the Split Trend graph of the WB approach currently shows signs of heavy traffic and significant split failures during the PM peak. This shows that the EB and WB approaches at this intersection are still heavily utilized even after the freeway interchange closure. Considering this issue, the option to slightly increase the green time to the EW approaches (taking green from the NS approaches) should be explored. Figure 10. La Cañada Dr. WB through performance after retiming Another operational issue identified is the SB left turn movement at this intersection which is also experiencing oversaturation and frequent split failures, during the AM and PM peak period as shown in Figure 11. It is suggested to increase the maximum green time for this movement to decrease split failures since the opposing NB through movement is performing below saturation. Figure 11. La Cañada Dr. SB left turn performance after retiming 6

8 Intersection of W Ina Rd. and N La Cholla Blvd. A one-week comparison of split failures using the Insights Split Trend graphs shows that after the freeway entrance was closed, and prior to the signal retiming, the NB left turn and EB through movements during the AM peak periods, and the NB through during the PM peak period, both declined in performance. The signal timing changes applied have managed to enhance the traffic conditions on the EB through and NB through movements. Some movements, as prior to the interchange closure, are experiencing heavy traffic and split failures which seem to be common across all critical movements. The decrease in the EB and WB volumes has provided the opportunity to increase the green time on the N and S approaches and ease the traffic conditions on these movements. Intersection of W Ina Rd. and N Shannon Rd. A one-week comparison of split failures using the TrafficLink Split Trend graphs reveals that only slight changes have occurred in the performance of this intersection as a direct result of the freeway interchange closure. However, the changes in the drivers route choice behavior due to the interchange closure along with the signal timing changes, has improved the performance of this intersection overall, especially during the PM peak period on the NB through, WB through, EB left turn, and SB left turn movements. Before the interchange closure, the WB through movement experienced the worst performance compared to other movements at the intersection in terms of split failures which a combination of volume reduction as a result of the interchange closure and the signal retiming (i.e., cycle length increase) has overall improved the traffic conditions at this intersection as shown in Figure 10. Figure 12. North Shannon Rd. WB through intersection performance comparison for all three time periods, bottom to top before closure, post closure, and post-retiming respectively 7

9 Although the traffic conditions on the WB through movement have improved after the signal retiming, this approach is still experiencing high split failures operating at or beyond saturation during the PM peak hours. A quick look at the NS split trend graphs, as shown in Figure 13, reveals that the NS movements are experiencing undersaturated performance, while the WB movement is failing to serve its demand in many cycles. If the goal is to more effectively distribute the congestion, then a slight redistribution of green time may help the EW movements. If the goal is to provide priority to the NS movements then the current setting is performing as intended. Figure 13. North Shannon Rd. SB and NB split trend performance after retiming Intersection of W Ina Rd. and N Camino de la Tierra The combination of volume changes (due to the interchange closure), and signal retiming has improved the performance of the EB left turn, and WB through movements at this intersection, which had been experiencing performance issues prior to the closure. As shown in Figure 14, the WB through movement was experiencing traffic performance issues prior to the closure which has been addressed after the interchange closure, and the signal retiming. Figure 14. N Camino de la Tierra performance comparison for before, and after the closure 8

10 Figure 15. N Camino de la Tierra signal performance after signal retiming While the signal retiming has improved the traffic conditions on all movements overall, the NB through movement at this intersection is still experiencing split failures during the PM peak period as shown in figure 15. However, the EB and WB movements show excess capacity in which it is expected that a redistribution of a few seconds of green time from the EW movements to the NS movements can prevent split failures from occurring. Figure 16. N Camino de la Tierra potential performance improvement opportunity on the NB through movement 9

11 Travel time evaluation Considering that Miovision TrafficLink is only installed on the W Ina Rd. corridor, it is only possible to evaluate the changes in travel time on the EW movements, and not for the NS movements. Figure 17 compares the travel times of the before period with the post-signal retiming period for the WB direction of the corridor, and Figure 18 shows this comparison for the EB direction. Figure 17. Travel time comparison from La Cañada Dr. to N Camino de la Tierra Figure 18. Travel time comparison from N Camino de la Tierra to La Cañada Dr. 10

12 Progression quality TrafficLink Insights uses the Arrivals on Green and Red to quantify the quality of progression in corridors. This information is available for both the NS and EW approaches at each intersection under evaluation. To prevent a lengthy report one sample of this data is presented here. Figure 19 shows the Arrivals on Green and Red of the NB approach at the intersection of La Cañada Dr. and W Ina Rd. before and after the retiming. As can be seen, the offsets on the NB movement has provided a good quality of progression with more than 70% of Arrivals on Green during the PM peak. As expected, due to poor coordination prior to retiming, a significant portion of vehicles experienced poor progression on the NB movement. Figure 19. Arrivals on Red and Green before and after retiming 11

13 Conclusion Miovision TrafficLink provided an evaluation of the signal performance of four intersections along the corridor of W Ina Dr. before and after a nearby freeway interchange closure and subsequent signal retiming. Using this high-resolution, signal performance data, it was identified that the corridor under study was experiencing performance issues even prior to the interchange closure. Additionally, the signal timing changes especially the increase in the cycle length has successfully mitigated a number of these performance issues. Overall the data suggests that drivers are still heavily using the EW approaches of the intersections of La Cañada Dr./W Ina Rd. and La Cholla Blvd./W Ina Rd. as opposed to the other two intersections which experienced more significant changes in travel patterns. Options around slight redistribution of green time at these intersections have also been explored in the report. This study also demonstrates how TrafficLink reports can be used to identify movements currently experiencing performance issues, and propose adjustments to the signal timings that can achieve optimal performance. Overall, the increase in cycle length to 150 seconds has mitigated many performance issues without significantly affecting the travel times on the W Ina Rd. corridor. For more information and to view webinars about Miovision TrafficLink, visit us online at miovision.com or call