Fabrizio Biora (MIZAR) and D. Pietrantonio (Comune di Torino)
CONTENTS
2 ANALYSIS OF THE LEEDS FIELD TRIAL
2.1 INTRODUCTION
2.2 BUS TRAVEL TIMES AND DELAY
2.2.1 Data collected
2.2.2 Results
2.2.3 Conclusions
2.3.1 Data collected
2.3.2 Results
2.3.3 Conclusions
2.4 CONFLICT STUDIES
2.4.1 Data collected
2.4.2 Results
2.4.3 Conclusions
2.5 PEDESTRIAN DELAY
2.5.1 Data collected
2.5.2 Results
2.5.3 Conclusions
2.6 SPEED PROFILES
2.6.1 Data collected
2.6.2 Results
2.6.3 Conclusions
2.7 QUEUE SURVEYS
2.7.1 Data collected
2.7.2 Results
2.7.3 Conclusions
2.8 COMPARISON WITH THE SIMULATION RESULTS
3 ANALYSIS OF THE TURIN FIELD TRIAL
3.1INTRODUCTION
3.2 TRAFFIC SCENARIO
3.3 CYCLE TIMES
3.3.1 Data collected
3.3.2 Results
3.3.3 Conclusions
3.4 AR JOURNEY TIMES
3.4.1 Data collected
3.4.2Results
3.4.3 Conclusions
3.5.1 Data collected
3.5.2 Results
3.5.3 Conclusions
3.6 STOPS
3.6.1 Data collected
3.6.2 Results
3.6.3 Conclusions
3.7 QUEUES
3.7.1 Data collected
3.7.2 Results
3.7.3 Conclusions
3.8 COMPARISON WITH THE SIMULATION RESULTS
APPENDIX A: ANALYSIS OF LEEDS MOVING OBSERVER TRAVEL TIME DATA
APPENDIX B: TABLE OF LINK TRAVEL TIMES
APPENDIX C: TABLE OF PERCENTAGE CHANGES ON LINK JOURNEY TIME FOR BUSES OF SERVICE 2
The trials used networks operating under the following conditions:
(a) Baseline conditions, Leeds and Turin
(b) SPOT operating alone, Leeds
(c) SPOT operating with the strategies devised for PRIMAVERA, Leeds and Turin
(d) SCOOT operating alone, Leeds
(e) SCOOT operating with the strategies devised for PRIMAVERA, Leeds
The strategies for each city are as follows:
Leeds
SCOOT - Starting and stopping waves, bus priority and speed advice.
SPOT - Horizontal queue model, bus priority and speed advice.
Details of the analysis of the Leeds results are given in Section 2. The initial analysis of the data highlighted a problem. It became apparent that the sample sizes in the conflict studies were not large enough to give statistically significant results. Additional funding was requested to collect more data, but this was not forthcoming. Therefore no conclusions can be drawn from the field trials about any changes in safety due to the implementation of the strategies.
The main improvements deduced due to adoption of the PRIMAVERA strategies in Leeds are:
SPOT - Cooperative auto-gating and speed advice.
Positive results on all the measured indicators have been obtained. Details of the analysis are given in Section 3. Section 3.1 presents an introduction to the field trials. A summary of the cases tested is given together with assumptions and references for the rest of the document. Sections 3.2 to 3.7 report the analysis of the data collected. Section 3.8 contains a comparison between the simulation and field trial results for the implemented strategy. Improvements produced from the adoption of the control strategy devised for PRIMAVERA in Turin can be summarised as follows:
Figure 2.1: The Dewsbury Road Trial Site
The Leeds field trial site and the data collection points are shown in Figure 2.1. The urban arterial chosen for the field trial in Leeds was a 3km section of the A653; the Dewsbury Road. This is one of the main radial routes into Leeds, carrying approximately 23,000 vehicles per day. It is also a heavily used public transport corridor, peak flows being in excess of 36 buses per hour. It is typical of many urban arterials in that it attempts to combine the functions of general traffic movement, public transport corridor, shopping street and residential area. On the chosen section there are ten signalised intersections including three pelican crossings. In terms of physical characteristics Dewsbury Road can be divided into three distinct parts. The outer section comprises largely a four lane carriageway with closely spaced pedestrian refuges, so that it is almost a dual carriageway. Fronting the road is a mixture of light industrial and residential uses. The central section passes through a local shopping centre with shops on both sides of the road. This part of Dewsbury Road comprises two traffic lanes with localised widening to three lanes at the two signal controlled junctions. The inner section, closest to the city centre, is made up of three lanes with one lane for city bound traffic and two lanes outbound, one which is reserved as an evening peak bus lane.
Six types of survey were carried out as follows:
Four of the major bus operators; Yorkshire Rider, Yorkshire Buses, Yorkshire Traction and West Riding Buses, equipped a total of 300 of their buses with transponders. The surveys during the field trials indicate that this resulted in approximately 65% of the buses using the Dewsbury Road being equipped.
Total bus journey times were measured by moving observers on buses. These times were measured for both inbound and outbound journeys over the Dewsbury Road trial site. Data was collected during both the morning and evening peaks.
| Strategy | No of matches | Journey Time (seconds) | s.d. | % change in time | % confidence |
|---|---|---|---|---|---|
| Baseline | 87 | 406 | 87 | ||
| SPOT + | 52 | 382 | 84 | -5.9 | 89 |
| SPOT alone | 48 | 455 | 118 | 12.1 | 99 |
| SCOOT + | 46 | 407 | 95 | 0.2 | 5 |
| SCOOT alone | 45 | 431 | 113 | 6.2 | 80 |
The first point to note is that the use of both SCOOT and SPOT on their own without the bus priority or traffic calming components increases bus travel times. This is because on this arterial the UTC systems impose strong co-ordination (green waves) for cars between the signals. Buses travel at different speeds to cars and therefore can drop out of this co-ordination. The bus journey time variability, as indicated by the standard deviation, also increases when the UTC systems are used on their own.
The introduction of the integrated strategies appears to reduce bus travel times back to baseline values when used with SCOOT and improves on the baseline values when used with SPOT. When comparing the integrated strategies against the UTC systems on their own, the integrated SPOT based strategy reduces the travel time by 16% when compared with SPOT alone, with virtually 100% confidence of a statistically significant change. The integrated SCOOT based strategy reduces the travel time by 6% with 72% confidence.
The number plate matching surveys do not distinguish between transponder equipped and non-equipped buses, for this the moving observer surveys need to be examined. The results for the entire bus fleet are shown in Table 2.2.
Inbound am Outbound am Inbound pm Outbound pm Time Time Time Time Baseline Number of Trips 35 32 26 29 Mean Time 377 353 322 357 s.d. 86 79 98 74 SPOT Number of Trips 38 37 41 36 Mean Time 413 345 311 356 s.d. 114 56 73 86 % change in time 9.5 -2.3 -3.4 -0.3 % confidence 87 37 40 4 Integrated SPOT Number of Trips 27 27 41 40 Mean Time 371 326 282 334 s.d. 88 47 69 77 % change in time -1.6 -7.6 -12.4 -6.4 % confidence 21 89 92 78 SCOOT Number of Trips 9 7 9 8 Mean Time 379 393 319 371 s.d. 58 69 47 43 % change in time 0.6 11.3 -0.9 3.9 % confidence 5 78 10 39 Integrated SCOOT Number of Trips 15 16 25 25 Mean Time 339 365 327 371 s.d. 80 100 67 76 % change in time -9.9 3.4 1.4 3.9 % confidence 85 35 17 50Table 2.2: Changes in bus travel time (Moving Observers)
The priority system only gives priority to transponder fitted buses travelling inbound. Table 2.3 shows the results for these equipped buses in this direction when the integrated strategies with bus priority are in operation.
Inbound am Inbound pm Time Time Integrated SPOT Number of Trips 21 25 Mean Time 348 290 s.d. 81 52 % change in time -7.7 -9.9 % confidence 78 85 Integrated SCOOT Number of Trips 11 12 Mean Time 348 331 s.d. 87 63 % change in time -7.7 2.8 % confidence 66 23Table 2.3: Changes in bus travel time (Transponder fitted buses)
The moving observer results indicate that during the AM peak both the SPOT and SCOOT based integrated strategies manage to reduce bus journey times for inbound buses by about 8% for the transponder equipped buses, when compared with the baseline.
The moving observer surveys also indicate that the integrated strategies reduce travel times for outbound buses during the AM peak, when compared to the UTC systems on their own. In the PM peak SPOT alone reduces travel times by 3.4% inbound with little change to outbound travel times. The integrated SPOT based strategy reduces inbound travel times by 12.4% and the outbound travel times by 6.4%. SCOOT is not helpful to buses during the PM peak, making little difference to inbound travel times and increasing outbound travel times by 3.9%, whether operating alone or with the integrated components.
Approximate total travel time savings for introducing the integrated components into the UTC systems are shown in Table 2.4. These are based on bus patronage figures during the two hour peaks and the moving observer travel time surveys.
AM Peak PM Peak Inbound Outbound Inbound Outbound Patronage 1700 800 1200 1500 Integrated SPOT Time saving (s) 42 19 29 22 Total saving (s) 71400 15200 34800 33000 Integrated SCOOT Time saving (s) 40 28 -8 0 Total saving (s) 68000 22400 -9600 0Table 2.4: Total bus travel time savings
The integrated SPOT based strategy produced a reduction in both bus journey times and journey time variability according to both the number plate matching and the moving observer surveys.
In addition a registration plate survey was undertaken between 0730 and 0930 using five timing points on the Dewsbury Road, inbound towards the city centre, on one day for each condition. The timing points can be seen in Figure 2.1. Unfortunately, a communications link failed on the day of the survey for the integrated SPOT strategy, which invalidated the results.
Figure 2.2: The Journey Time route links
Table 2.5 shows the changes in travel time on each link within these
moving observer routes. The travel times for each of the integrated strategies
is compared with the travel time for the UTC system operating without the
integrated components. Therefore the integrated SCOOT strategy is compared
against SCOOT without the starting and stopping wave, bus priority and
traffic calming components. The integrated SPOT strategy is likewise compared
against SPOT without the horizontal queue model, bus priority and traffic
calming components
The typical flows on each link during the two hour peak periods are also shown. These have been used to calculate the overall change in travel time for all the vehicles using the network covered by the moving observers. These changes in overall network time are given below the main table.
AM Peak AM Peak PM Peak PM Peak % change in % change in % change in % change in Length AM Peak Travel Time Travel Time PM Peak Travel Time Travel Time (m) Flow SPOT + SCOOT + Flow SPOT + SCOOT + Link 1 288 1200 -0.82 3.92 1400 -6.70 9.06 2 235 1200 -16.84 -0.19 1400 -7.00 15.67 3 153 1200 -22.26 3.18 1300 -3.14 -12.29 4 204 1200 8.12 -9.97 1300 4.84 -7.62 5 401 1200 2.26 -5.75 2000 -0.98 3.47 6 395 1200 5.92 -7.35 2150 -4.23 1.06 7 193 1200 -9.06 -7.27 2150 -9.33 5.83 8 493 1200 -1.39 8.59 2100 13.81 54.28 9 246 1350 -11.43 7.12 1550 -8.58 5.31 10 451 800 8.44 5.70 800 1.17 7.84 11 552 1250 -22.79 22.48 800 -41.54 99.34 12 507 2200 11.86 -1.79 1900 3.90 12.29 13 180 2200 24.85 13.25 2100 -16.90 -4.50 14 390 2200 4.49 5.10 2100 1.45 10.97 15 405 2200 11.22 6.02 2500 1.79 -16.68 16 492 700 -0.18 -6.47 1050 9.24 -10.18 17 189 2000 -6.90 2.16 2500 3.30 23.94 18 231 2000 1.09 4.59 2500 2.99 -2.71 19 187 2000 -6.23 9.64 1400 6.35 81.44 20 256 900 29.86 34.60 700 6.42 -30.36 21 248 1700 18.12 -2.95 1200 12.91 -14.73 22 290 1700 -10.26 -9.54 1200 5.44 36.93 23 127 1600 -6.90 -3.78 1450 16.03 -24.30 24 226 1600 -13.37 18.34 1450 30.51 -6.71 25 208 1950 -8.48 4.79 1900 40.13 -22.60 26 469 700 7.76 13.53 1000 37.77 8.07 27 494 1100 23.65 -12.85 1000 7.19 2.07 28 246 900 -8.81 22.24 1300 6.78 10.62 29 67 800 2.82 -11.01 800 -3.59 2.25 Network 3056 81343 Network 125755 136121 change(s) change(s)Table 2.5: The changes in travel times for each link
The total change in travel times for all vehicles travelling on these links, assuming a vehicle occupancy of 1.4 per vehicle is shown at the bottom of Table 2.5. For the AM Peak, these increases in car travel time are easily countered by the decrease in bus travel times in the same period. (See Table 2.4) For the PM Peak the increased car travel times exceed the bus travel time savings. The extra delay to cars during the PM Peak is probably a result of using the bus priority component. This component is only giving priority towards the city centre and during the PM Peak this is against the main traffic flow. The staging arrangements at some of the bus priority junctions means that extending green times for buses inbound can result in an increase in red time for vehicles travelling outbound. It is therefore recommended that the bus priority component is not used during the PM Peak.
(1) Integrated SPOT based strategy (week beginning 25th of July 1994)
(2) Baseline studies (week beginning 1st of August 1994)
(3) Integrated SCOOT based strategy (week beginning 25th of October 1994)
| Conflict Type | Vehicle-Vehicle | Vehicle-Pedestrian | Total |
|---|---|---|---|
| Serious conflicts | 2 | 1 | 3 |
| Border line cases | 1 | 0 | 1 |
| Slight conflicts | 11 | 11 | 22 |
| Total | 14 | 12 | 26 |
(b) Baseline conditions
| Conflict Type | Vehicle-Vehicle | Vehicle-Pedestrian | Total |
|---|---|---|---|
| Serious conflicts | 1 | 1 | 2 |
| Border line cases | 1 | 2 | 3 |
| Slight conflicts | 2 | 5 | 7 |
| Total | 4 | 8 | 12 |
(c) Integrated SCOOT based strategy
| Conflict Type | Vehicle-Vehicle | Vehicle-Pedestrian | Total |
|---|---|---|---|
| Serious conflicts | 0 | 0 | 0 |
| Border line cases | 1 | 0 | 1 |
| Slight conflicts | 12 | 3 | 15 |
| Total | 13 | 3 | 16 |
Using the statistical procedures of Nicholson (1987)[1] but substituting days of observed conflicts instead of years of observed accidents, we can say with 95% confidence that there has been a significant change from the baseline rate if the total number of all conflicts is outside the range 3 to 21. Similarly if the total number of serious conflicts is greater than 5 we can say with 95% confidence that there has been a significant increase. Thus none of the observed changes were significant, except for the total number of conflicts under the SPOT based strategy.
Implementation Site Number Baseline 1 2/3 4 5 6 No. of pedestrians 379 206 719 241 451 Average Delay (s) 17.44 13.88 17.10 19.10 16.45 SPOT No. of pedestrians 286 247 241 172 362 Average Delay (s) 28.08 11.22 28.11 27.72 28.67 Delta Average Delay 10.64 -2.66 11.01 8.62 12.22 % change 61.01 -19.17 64.37 45.11 74.25 % confidence 100 93 100 100 100 Integrated SPOT No. of pedestrians 418 247 344 228 365 Average Delay (s) 31.91 11.63 34.18 35.17 25.29 Delta Average Delay 14.47 -2.26 17.08 16.07 8.84 % change 82.97 -16.25 99.90 84.11 53.74 % confidence 100 86 100 100 100 SCOOT No. of pedestrians 437 249 404 237 314 Average Delay(s) 28.63 11.56 25.51 32.78 34.80 Delta Average Delay 11.19 -2.33 8.41 13.68 18.35 % change 64.17 -16.75 49.21 71.61 111.52 % confidence 100 88 100 100 100 Integrated SCOOT No. of pedestrians 369 105 404 235 279 Average Delay (s) 23.91 16.46 24.92 24.73 30.72 Delta Average Delay 6.47 2.57 7.82 5.62 14.27 % change 37.08 18.54 45.73 29.44 86.72 % confidence 100 40 100 100 100
If the average pedestrian flows are used at each site, it is possible to estimate the total change in delay across the five sites. This is shown in the following table.
Table 2.10: Change in Total Pedestrian Delay
Site No. 1 2/3 4 5 6 Total Average flow 378 211 422 223 354 1588 SPOT Delay change (s) 4020 -561 4650 1918 4327 14354 Integrated SPOT Delay change (s) 5467 -476 7216 3577 3131 18915 SCOOT Delay change (s) 4228 -490 3554 3045 6499 16836 Integrated SCOOT Delay change (s) 2443 543 3303 1252 5054 12595All the systems result in increase in total pedestrian delay. This is not surprising as the UTC systems make full use of the available road space, resulting in fewer gaps for pedestrians to cross the road. However, if these increases in pedestrian delay are compared with the savings in travel time from implementing the strategies at this trial site, the bus travel time savings are an order of magnitude larger than the increases in pedestrian delay.
The comparison between the UTC systems on their own and the integrated strategies is seen in Table 2.11.
Average % change % change(s) confidence Integrated SPOT +3.18 +12.7 100 Integrated SCOOT -1.91 -7.1 97Table 2.11: Changes in Pedestrian Delay
There is an increase in pedestrian delay when the integrated SPOT based strategy is compared against SPOT alone. There is a slight decrease for the integrated SCOOT based strategy.
It is interesting to note that the only site which shows any improvement in pedestrian delay is the one just downstream of the VMS.
Click here for Figure 2.3
Figure 2.3: The location of the speed counters
The data presented here covers five complete days (24 hours) of data for a period before the installation compared with a corresponding five day period after the installation.
Click here for Figure 2.4
Figure 2.4: The Speed Profile at S12
Click here for Figure 2.5
Figure 2.5: The Speed Profile at S13
Click here for Figure 2.6
Figure 2.6: The Speed Profile at S14
The changes in the percentage of vehicles exceeding the speed limit,
the mean speed and standard deviation of the speed are shown in Table 2.12.
| % speeding | Mean Speed (mph) | % confidence in change in Mean Speed | Standard Deviation | ||||
|---|---|---|---|---|---|---|---|
| Site | Before | After | Before | After | Before | After | |
| S12 | 18 | 3 | 35.3 | 31.3 | 100.0 | 6.6 | 5.8 |
| S13 | 16 | 3 | 31.0 | 29.2 | 100.0 | 6.6 | 4.9 |
| S14 | 6 | 5 | 29.4 | 29.2 | 99.9 | 6.6 | 6.5 |
Due to the very large sample size, there is a very high confidence that the mean speed of vehicles approaching the pelican crossing has decreased. A recent TRL report (Finch et. al. 1994) has indicated that a 1 mph reduction in vehicle speeds can result in a 5% reduction in accident rates, therefore significant safety benefits are predicted for this site. There is even strong evidence that the influence of the VMS extends downstream beyond a pelican crossing as a small but statistically significant reduction in speed has been recorded here.
Click here for Figure 2.7
Figure 2.7: The queues at an approach to Beeston Ring Road junction
Click here for Figure 2.8
Figure 2.8: The queues at another approach to Beeston Ring Road junction
The average queues on each arm of the surveyed junctions during the
survey periods have been calculated. These have been added together to
give the total average queues on all arms of the junctions. The results
are presented in Tables 2.13 and 2.14. During the AM Peak, the integrated
SCOOT based strategy produces an increase of 8.4% in total queue length
when compared with SCOOT alone. The integrated SPOT based strategy increases
the total queue lengths by 11.7%. In terms of the extra visual intrusion
produced by this increase in queue lengths, this is approximately equivalent
to adding one extra car to the end of each queue during the peak period.
AM Peak Baseline SCOOT SCOOT+ SPOT SPOT+ Tommy Wass 66 62 69 83 95 Westland Road 21 17 21 19 18 Middleton Grove 12 17 22 15 25 Parkside / Garnet 32 37 36 35 31 Dewsbury / Tunstall 20 23 21 29 31 Hunslet Hall Road 20 16 24 26 29 Tunstall / Garnet 32 33 29 31 36 Total 204 205 222 238 266 % change 8.36 11.67Table 2.13: The total average queues on all arms (AM Peak)
PM Peak Baseline SCOOT SCOOT+ SPOT SPOT+ Tommy Wass 86 77 88 91 82 Westland Road 16 17 25 19 16 Middleton Grove 13 20 20 17 28 Parkside / Garnet 37 35 38 29 36 Dewsbury / Tunstall 26 11 29 20 40 Hunslet Hall Road 23 14 19 14 16 Tunstall / Garnet 31 39 39 31 30 Total 232 214 257 220 249 % change 20.45 12.96Table 2.14: The total average queues on all arms (PM Peak)
SPOT Simulated Actual SCOOT Simulated Actual Link % change % change Link % change % change 1 2.7 -0.8 1 -0.8 3.9 2 2.9 -16.8 2 -9.9 -0.2 3 -1.0 -22.3 3 -8.3 3.2 4 -2.8 8.1 4 0.0 -10.0 5 -4.6 2.3 5 31.1 -5.8 6 -3.1 5.9 6 -1.8 -7.4 7 -7.7 -9.1 7 24.0 -7.3 8 -2.0 -1.4 8 11.3 8.6 9 -1.8 -11.4 9 10.3 7.1 10 0.1 8.4 10 0.0 5.7 11 -0.5 -22.8 11 11.2 22.5 12 7.1 11.9 12 -12.4 -1.8 13 3.4 24.9 13 -3.6 13.3 14 -7.5 4.5 14 -2.5 5.1 15 -10.6 11.2 15 0.0 6.0 16 -6.4 -0.2 16 7.3 -6.5 17 -0.2 -6.9 17 0.0 2.2 19 -25.9 -6.2 19 18.3 9.6 20 2.7 29.9 20 5.1 34.6 21 -0.3 18.1 21 0.0 -3.0 22 -1.0 -10.3 22 4.9 -9.5 23 -3.2 -6.9 23 6.3 -3.8 24 3.4 -13.4 24 0.2 18.3 25 8.3 -8.5 25 2.0 4.8 26 -0.3 7.8 26 9.2 13.5 27 -0.9 23.7 27 -1.0 -12.9 28 1.8 -8.8 28 43.2 22.2 Total -4.2 0.1 Total 1.7 3.7 change changeTable 2.15: Simulated vs actual changes in travel times
As can be seen there is rarely agreement between the simulated and observed results. There are a number of reasons for this. Firstly there is the usual problem with field trials of not knowing whether the changes being measured are due entirely to the strategy or to some external influence such as the weather or fluctuations in demand. Similarly, simplifications had to be made to the network and signal plan representations in the simulations which could result in discrepancies. Secondly, the simulations highlighted some areas where problems could occur. With these in mind the systems implemented on-street were changed to try and reduce these problems. For example, the simulations revealed that when using SCOOT, constraining the whole area to an 88s upper limit on cycle time did not allow cars to benefit as much as possible. Therefore during the field trials the area was split into two regions, one with the 88s constraint the other without. The simulations and the field trials are thus comparing different systems. Overall, for the whole network, the SCOOT results show reasonable agreement, both showing a small increase in travel time. The SPOT results are not so good, the simulations predicting a reduction in travel time, the field trials showing little change.
The field trials have been designed to see whether improvements as predicted by simulations in Deliverable 12: "Evaluation of Simulated Strategies", occur in reality. Statistical tests have been carried out on the data collected during the field trials. These are used to produce confidence levels that the measured changes are real and not due simply to inherent variability of the data.
The trials in Torino cover the network operating under the following conditions:
SPOT+ = SPOT + Cooperative Auto-gating + Speed Advice
The strategy adopted is totally decentralised and no interventions have been allowed by the Area Level Control of the UTOPIA system in Torino. The Traffic Control Centre has been used merely for its monitoring functionality.
Field trials have covered the Corso Grosseto area. The test site includes
seven controlled intersections located along the Corso Grosseto arterial.
See Deliverable 4: "Description of the Test Sites" for a more detailed
description of the test site. For simplicity, in the rest of the document
the intersections are identified by a numeric index. The following table
shows the correspondence between the numeric code and the intersections,
the diagram below a schema of the topology of the network:
| Code | Intersection |
|---|---|
| 11 | Corso Grosseto - Via Casteldelfino |
| 12 | Corso Grosseto - Via Fea |
| 13 | Corso Grosseto - Via Bibiana |
| 14 | Corso Grosseto - Via Chiesa della Salute |
| 15 | Corso Grosseto - Via Ala di Stura |
| 16 | Corso Grosseto - Corso Vercelli - Via Botticelli |
| 17 | Corso Vercelli - Via Toscanini - Via Porpora |
Click here for Figure 3.1
Figure 3.1: The topology of the Corso Grosseto test site
The baseline surveys were carried out from September to October 1994 while the SPOT+ surveys were carried out in June 1995. Between the Baseline and SPOT+ surveys significant modifications to one link, connecting intersections 15 and 16, have been implemented. These modifications have narrowed the carriageway from 4 to 3 lanes in order to facilitate the entrance and the exit from the motorway going to the airport. The changes have resulted in heavy congestion on the link. The link has therefore not been considered in the evaluation of the benefits/disbenefits introduced by the system.
Results are presented for the AM and PM peak periods. Timings for the peaks are 7:00-9:30 for AM and 17:00-19:30 for PM.
The following series of diagrams represent the topology, stage planning and turning movements for intersections 11, 15 and 16 where Manual Classified Counts took place. Intersections 12, 13, 14 have a topology very similar to intersection 11 and the same stage planning.
Click here for Figure 3.2.
Figure 3.2: Topology and stage planning for intersection 11
Click here for Figure 3.3.
Figure 3.3: Topology and stage planning for intersection 15
Click here for Figure 3.4.
Figure 3.4: Topology and stage planning for intersection 16
The MCC surveys also gave an indication about the traffic composition
on the arterial as summarised in the following table.
| Category | Percentage |
|---|---|
| Cars | 93.8 % |
| Lorries | 5.6 % |
| Heavy Goods Vehicles | 0.6 % |
More than 30 ATC counters were deployed in the Corso Grosseto area. For each day the flow profile, sampled at 5 min. intervals has been recorded. The following diagrams show the typical flow profile on one inbound and one outbound section. From the diagrams the two peaks can be easily identified. In absolute terms the dominant flow in the AM peak is in the inbound direction while during the PM peak inbound and outbound flows are comparable.
Click here for Figure 3.5.
Figure 3.5: Flow profile on Corso Grosseto, inbound direction
Click here for Figure 3.6.
Figure 3.6: Flow profile on Corso Grosseto, outbound direction
For the assessment of the control strategies flows on all the links are needed. The following table and figures show the flows on each link aggregated for the peak periods. Two detectors were placed on links 15-16 and 16-15, one just at the exit of the upstream junction and the other after the confluence of the motorway coming and going to the airport.
Flows on the side roads represent only a small percentage of the flow
on the arterial except for Via Botticelli and Corso Vercelli where they
are comparable. These two roads are arms of intersection 16 which is the
most saturated in the field trial area.
Link AM: 7:00-9:30 PM: 17:00-19:30 entry 11 3882 4328 11 to 12 4289 4175 12 to 13 4717 4684 13 to 14 4340 4353 14 to 15 4021 4271 15 to 16, 1 3965 4188 15 to 16, 2 4769 4375 16 to 17 3341 2659 exit 17 2247 2337 entry 17 1866 2101 17 to 16 3434 3459 16 to 15, 1 5551 5290 15 to 15, 2 6576 7352 15 to 14 3911 4678 14 to 13 4222 4984 13 to 12 4913 5410 12 to 11 4683 5498 exit 11 4628 5014 Via Casteldelfino 1423 1407 Via Fea 448 821 Via Bibiana 1110 1270 Via Chiesa della Salute 590 994 Via Ala Nord 962 1523 Via Ala Sud 1248 1396 Corso Vercelli Sud 3572 3597 Via Botticelli 3909 3877 Via Toscanini 1509 1822 Via Porpora 826 1146Table 3.3: Flows during the peaks on the links
Flows on the two routes, A and B, covered by the moving observers are now presented, along with the flows on the side roads.
Click here for Figure 3.7.
Figure 3.7: Flows on Route A, AM and PM peaks
Click here for Figure 3.8.
Figure 3.8: Flows on Route B, AM and PM peaks
Click here for Figure 3.9.
Figure 3.9: Flows on the side roads, AM and PM peaks
JUNCTION 11 Stage 1 2 3 CYCLE Baseline 45 16 29 90 SPOT+ AM Av. 69 18 26 113 SPOT+ PM Av. 67 18 31 116 SPOT+ AM St.Dev. 14 0 9 14 SPOT+ PM St.Dev. 14 0 9 16Table 3.4: Cycle timings for intersection 11
JUNCTION 12 Stage 1 2 3 CYCLE Baseline 47 14 29 90 SPOT+ AM Av. 75 22 15 112 SPOT+ PM Av. 72 21 18 111 SPOT+ AM St.Dev. 12 6 2 14 SPOT+ PM St.Dev. 15 6 4 15Table 3.5: Cycle timings for intersection 12
JUNCTION 13 Stage 1 2 3 CYCLE Baseline 47 16 29 90 SPOT+ AM Av. 65 15 33 113 SPOT+ PM Av. 62 15 36 113 SPOT+ AM St.Dev. 15 0 10 13 SPOT+ PM St.Dev. 22 0 9 21Table 3.6: Cycle timings for intersection 13
JUNCTION 14 Stage 1 2 3 CYCLE Baseline 40 16 34 90 SPOT+ AM Av. 83 15 17 115 SPOT+ PM Av. 74 15 30 119 SPOT+ AM St.Dev. 8 0 5 7 SPOT+ PM St.Dev. 17 0 11 16Table 3.7: Cycle timings for intersection 14
JUNCTION 15 Stage 1 2 3 CYCLE Baseline 49 16 25 90 SPOT+ AM Av. 66 26 19 111 SPOT+ PM Av. 59 29 28 117 SPOT+ AM St.Dev. 13 4 5 15 SPOT+ PM St.Dev. 13 2 9 15Table 3.8: Cycle timings for intersection 15
JUNCTION 16 Stage 1 2 3 4 5 6 CYCLE Baseline 25 17 18 13 23 14 110 SPOT+ AM Av. 40 21 33 27 14 27 162 SPOT+ PM Av. 40 22 33 29 23 32 179 SPOT+ AM St.Dev. 3 8 2 3 8 4 13 SPOT+ PM St.Dev. 3 8 2 2 10 2 11Table 3.9: Cycle timings for intersection 16
JUNCTION 17 Stage 1 2 3 CYCLE Baseline 20 49 41 110 SPOT+ AM Av. 15 28 46 89 SPOT+ PM Av. 17 40 45 112 SPOT+ AM St.Dev. 1 7 17 16 SPOT+ PM St.Dev. 4 13 16 22Table 3.10: Cycle timings for intersection 17
D=sR2/2(s-q)C
where:
s: is the saturation flow
q: is the demand on the link
R: is the red length in the cycle
C: is the cycle length
The above formula is valid as long as (i) the distribution arrival pattern for the link is uniform and (ii) the degree of saturation is less than approximately 0.7. These conditions are valid for the side roads as (i) there are no other intersections close to the controlled ones and (ii) the degree of saturation on the side roads is low enough.
First an analysis of link journey time is presented. As described in the introduction journey times on link 15 to 16 have been excluded by the evaluation.
Besides significant reductions in journey time on most of the links, it should be noted that there is also a general reduction in the variability, shown by the reduction in the standard deviation of the distribution.
ROUTE A PERIOD AM entry 11 to 12 to 13 to 14 to 15 to 16 to 11 12 13 14 15 16 17 Baseline Trips 50 50 50 50 50 n.a. 50 Time (s) 27 52 39 25 42 n.a. 45 s.d. (s) 14 20 16 17 22 n.a. 14 speed (km/h) 27 21.5 12.8 24.5 21.6 n.a. 7.8 SPOT+ Trips 45 45 45 45 45 n.a. 65 Time (s) 25 32 20 18 40 n.a. 32 % change -5 -39 -48 -27 -6 -30 confidence 47 100 100 98 34 100 s.d. (s) 17 12 13 10 22 n.a. 21 speed (km/h) 28.6 35.5 24.7 33.7 23 n.a. 11.1
Table 3.11: Journey time on the links, Route A AM
ROUTE A PERIOD PM entry 11 to 12 to 13 to 14 to 15 to 16 to 11 12 13 14 15 16 17 Baseline Trips 50 50 50 50 50 n.a. 50 Time (s) 25 57 42 31 40 n.a. 61 s.d. (s) 12 19 19 20 25 n.a. 38 speed (km/h) 29.4 19.7 12 19.9 122.7 n.a. 5.8 SPOT+ Trips 32 32 32 32 32 n.a. 50 Time (s) 20 39 27 17 39 n.a. 36 % change -20 -32 -34 -46 -5 -41 confidence 90 100 100 100 15 10 s.d. (s) 15 16 13 4 19 n.a. 18 speed (km/h) 36.5 29 18.3 37 23.8 n.a. 9.8Table 3.12: Journey time on the links, Route A PM
ROUTE B PERIOD AM entry 17 to 16 to 15 to 14 to 13 to 12 to 17 16 15 14 13 12 11 Baseline Trips 50 50 50 50 50 50 50 Time (s) 55 128 69 42 46 43 44 s.d. (s) 18 65 29 28 23 18 29 speed (km/h) 11 2.7 42.1 21.7 13.3 11.7 25.4 SPOT+ Trips 65 45 82 45 45 45 45 Time (s) 62 138 75 34 21 13 35 % change 13 8 9 -21 -55 -69 -21 confidence 85 55 81 91 100 100 94 s.d. (s) 33 63 23 16 14 5 16 speed (km/h) 9.8 2.5 38.7 27.3 29.8 38 31.8Table 3.13: Journey time on the links, Route B AM
ROUTE B PERIOD PM entry 17 to 16 to 15 to 14 to 13 to 12 to 17 16 15 14 13 12 11 Baseline Trips 50 50 50 50 50 50 50 Time (s) 100 159 66 50 49 48 32 s.d. (s) 18 60 23 34 24 17 10 speed (km/h) 6.1 2.2 44.1 18.4 12.4 10.5 34.7 SPOT+ Trips 50 50 65 32 32 32 32 Time (s) 59 156 79 33 25 21 34 % change -41 -2 19 -34 -50 -56 6 confidence 100 19 100 100 100 100 51 s.d. (s) 27 62 20 16 18 10 14 speed (km/h) 10.3 2.2 37 27.8 24.8 23.6 32.7Table 3.14: Journey time on the links, Route B PM
Route Journey Times
The following table shows the total route journey time for routes A
and B in the AM and PM peaks. In all the periods a significant reduction
in journey times has been found.
Case Route A AM Route B AM Route A PM Route B PM Baseline (s) 231 428 255 504 SPOT+ (s) 167 379 177 407 % change -28 -12 -31 -19Table 3.15: Journey time on the routes
Click here for Figure 3.10.
Figure 3.10: Journey time on the routes
The results show how the new system improves coordination between intersections. Significant benefits can be found during both peaks on the two routes. Such a result due to the coordination on the arterial should have introduced additional delays on the side roads. The aim of the following section is to evaluate a travel time indicator of the controlled network that also consider the side roads.
Network Travel Time
The aim of this section is to produce a travel time indicator for the network. Travel time is defined as the journey time on the link multiplied by the flow travelling on the link during the observation period. An indicator has been calculated for each intersection and then aggregated across the network.The following tables report the intersection and network indicators. Please refer to Appendix B for a table containing the link travel times. Values are in vehicle seconds.
Intersection Baseline SPOT+ % change 11 361155 335222 -7.2 12 448386 224493 -49.9 13 417582 231184 -44.6 14 291362 243011 -16.6 15 702489 770377 9.6 16 638746 737090 15.4 17 341068 320183 -6.1 Network 3200789 2861559 -10.6Table 3.16: Travel time in the network, AM peak
Intersection Baseline SPOT+ % change 11 333310 342279 2.7 12 523511 320563 -38.8 13 485479 305781 -37 14 396649 274885 -30.7 15 707643 831559 17.51 16 748688 817841 9.2 17 488736 350454 -28.3 Network 3684016 3243361 -11.9Table 3.17: Travel time in the network, PM peak
Click here for Figure 3.11.
Figure 3.11: Travel time in the network, % changes
Hour 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Base 336 361 340 360 310 318 332 330 330 350 355 323 324 322 341 329 330 SPOT+ 336 336 330 325 328 322 320 323 323 327 333 351 342 329 324 316 314 % change 0.03 -6.84 -2.97 -9.70 6.04 1.38 -3.47 -2.15 -2.12 -6.49 -6.27 8.60 5.49 2.05 -5.01 -3.98 -4.84Table 3.18: Bus Journey Time, Outbound direction
Hour 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Base 278 278 277 269 294 278 273 280 276 270 287 285 273 271 250 243 241 SPOT+ 262 268 264 259 254 254 247 246 243 245 256 263 264 257 250 245 239 % change -5.86 -3.60 -4.63 -3.79 -13.5 -8.36 -9.36 -12.4 -12.0 -9.31 -10.8 -7.70 -3.30 -5.38 -0.08 0.78 -0.46Table 3.19: Bus Journey Time, Inbound direction
Click here for Figure 3.12.
Figure 3.12: Bus Journey Time, Outbound direction
Click here for Figure 3.13.
Figure 3.13: Bus Journey Time, Inbound direction
From the above diagrams it clear that for much of the day the inbound direction benefits from the new control strategy applied even if no bus priority measure were taken. This data confirms the private traffic data (journey time on route A) even if the effect is less marked. In the outbound direction benefits are shown in the AM peak and Off-peak periods while journey times are greater with the SPOT+ system during part of the PM peak.
The following table shows the average benefit/disbenefits introduced
during the two peaks and during the whole day.
Period 7-23 7-9 17-19 Baseline 335 346 334 SPOT+ 328 334 342 % change -2 -3.3 2.3Table 3.20: Bus Journey Time, Outbound direction
Period 7-23 7-9 17-19 Baseline 272 278 281 SPOT+ 254 265 261 % change -6.6 -4.7 -7.3Table 3.21: Bus Journey Time, Inbound direction
Click here for Figure 3.14.
Figure 3.14: Bus Journey Time, Outbound direction
Click here for Figure 3.15.
Figure 3.15: Bus Journey Time, Inbound direction
Stops have been recorded by moving observers on the route travelled. Thus data about stops cover only the links on the arterial.
By using the stopping rate for floating cars, the total number of vehicle stopped per peak period has been calculated by multiplying the measured rate by the flow on the link. By summing the stops on the links it is possible to work out an intersection and a network indicator.
ROUTE A PERIOD AM entry 11 to 12 to 13 to 14 to 15 to 16 to 11 12 13 14 15 16 17 Baseline Trips 50 50 50 50 50 n.a. 50 Stops 23 31 41 7 25 n.a. 47 Stop rate (%) 46 62 82 14 50 n.a. 94 SPOT+ Trips 45 45 45 45 45 n.a. 65 Stops 18 15 13 2 21 n.a. 46 Stop rate (%) 40 33 29 4 47 n.a. 71Table 3.22: Stops on the links, Route A AM
ROUTE A PERIOD PM entry 11 to 12 to 13 to 14 to 15 to 16 to 11 12 13 14 15 16 17 Baseline Trips 50 50 50 50 50 n.a. 50 Stops 33 39 42 11 20 n.a. 47 Stop rate (%) 66 78 84 22 40 n.a. 94 SPOT+ Trips 32 32 32 32 32 n.a. 50 Stops 9 17 18 0 14 n.a. 35 Stop rate (%) 28 53 56 0 44 n.a. 70Table 3.23: Stops on the links, Route A PM
ROUTE B PERIOD AM entry 17 to 16 to 15 to 14 to 13 to 12 to 17 16 15 14 13 12 11 Baseline Trips 50 50 50 50 50 50 50 Stops 40 49 29 17 33 48 14 Stop rate (%) 80 98 58 34 66 96 28 SPOT+ Trips 65 45 82 45 45 45 45 Stops 45 43 48 16 6 3 18 Stop rate (%) 70 98 59 36 13 7 40Table 3.24: Stops on the links, Route B AM
ROUTE B PERIOD PM entry 17 to 16 to 15 to 14 to 13 to 12 to 17 16 15 14 13 12 11 Baseline Trips 50 50 50 50 50 50 50 Stops 47 50 30 19 34 43 5 Stop rate (%) 94 100 60 38 68 86 10 SPOT+ Trips 50 50 65 32 32 32 32 Stops 27 50 45 10 6 9 9 Stop rate (%) 54 100 69 31 19 28 28Table 3.25: Stops on the links, Route B PM
Data on the links has been agregated at junction and arterial level using the flows on the links. Results are shown in the following table and diagram.
AM PM Inters Baseline SPOT % change Baseline SPOT % change 11 3097 3426 11% 3406 2764 -19% 12 7376 1757 -76% 7909 3739 -53% 13 6655 1926 -71% 7324 3569 -51% 14 1937 1584 -18% 2735 1462 -47% 15 5824 5726 -2% 5564 6351 14% 16 3365 3350 0% 3459 3459 0% 17 4633 3678 -21% 4474 3000 -33% Arterial 32887 21446 -35% 34871 24344 -30%Table 3.26: Stops at the intersections and global indicators
Click here for Figure 3.16.
Figure 3.16: Stops at the intersections and global indicators, % changes
The benefits may be quantified by a 35% reduction in the number of vehicle stopped during the AM peak and a 30% reduction during the PM peak.
These results only consider the arterial and not the side roads but, as no coordination can be provided on the side roads due to the distance of upstream signals, they are a significant indicator of the correct behaviour of the strategy applied.
Intersection 11 Grosseto Grosseto Casteldelfino Outb. 12 to 11 Inb. entry 11 AM SPOT+ 23 18.3 12.3 Baseline 32.3 18.0 10.9 % change -28.9 0 12.5 PM SPOT+ 24.3 17.65 12.2 Baseline 14.9 16 11.5 % change 63.2 7.4 6.6Table 3.27: Queues on intersection 11
As a consequence of the higher cycle time the average queue on the side roads has increased slightly in the PM peak and had a more marked increase during AM peak. Because of the low flow on the link, the changes have a very low absolute value, of the order of 1-2 vehicles. No major changes can be found on the arterial in the inbound direction, entry link for the controlled area, while the outbound direction passes from a marked reduction during AM to a significant increase during PM. This is because there is a different coordination on the arterial as the same behaviour can be found in the link journey time analysis (link 12 to 11 passes from a 21% decrease in AM to a 6% increase during PM). The absolute value for queue length however stays at a low level considering that Corso Grosseto is a four lane carriageway.
Intersection 15 Grosseto Grosseto Ala Sud Ala Nord Outb. 16 to 15 Inb. 14 to 15 AM SPOT+ 26.9 23.7 9.4 14.2 Baseline 45.8 29.6 8.8 12.6 % change -41 -20 6.6 12.4 PM SPOT+ 34.3 28.7 19.8 15.3 Baseline 44.9 16.4 19.9 11.7 % change -23.6 21.9 0 30.4Table 3.28: Queues on intersection 15
Again because of the higher cycle times, queues on the side roads increase slightly, but not greater than 3 vehicles in absolute terms. A strong reduction in queues in the outbound direction of Corso Grosseto is noted, probably due to better management of the left turning traffic on Via Ala di Stura Sud. Changes in the coordination of the intersections, as for junction 11, result in the benefits of the AM becoming disbenefits in PM. Unlike junction 11 the changes in the average queues are not reflected in the journey times on the link.
Intersection 16 Grosseto 15 Vercelli Sud Vercelli Botticelli to 16 Nord 17 to 16 AM SPOT+ n.a. 34.8 38 46.4 Baseline n.a. 19.9 38.5 30.7 % change 74.7 -1.4 51.2 PM SPOT+ n.a. 40.3 60.25 60.6 Baseline n.a. 36.8 58.3 69.3 % change 9.5 3.2 -12.5Table 3.29: Queues on intersection 16
Intersection 16 is the place where major changes in queues have been detected. Major changes occurred on the two side roads where, with the new control strategy, significant increases in the average queue length can be seen in the morning peak. No relevant changes have been measured in the internal link between junction 17 and 16. For the absolute values of the changes it should be noted that all the links of this intersection have four lanes so even the biggest change gives only an increase of 4 vehicles per lane on Via Botticelli during the AM peak. It should be noted that link 15 to 16 has been excluded by this evaluation due to the problems described in the introduction.
Intersection 17 Vercelli Toscanini Vercelli Sud Porpora Nord entry 17 16 to 17 AM SPOT+ 8.5 12 12.9 5 Baseline 13 12.4 13.1 5.9 % change -34.5 -3.3 -1 -15 PM SPOT+ 12.5 11.3 13.3 6.5 Baseline 15.5 10.7 12.8 7.2 % change -19 4.7 3.9 -9Table 3.30: Queues on intersection 17
Significant reduction of the queues occurred on Corso Vercelli Nord and Via Porpora during both peaks. No significant changes can be found on the other links. Absolute changes are not large, the biggest is of 5 vehicles on the four lanes of Corso Vercelli.
Even if significant in percentage terms, absolute changes, both for reductions and increases, have been limited to less than 4 vehicles per lane.
The predicted change in network travel time agrees well with the field trials. The field trials indicate a 10% reduction against the 9% reduction predicted by the simulation results.
For bus journey time again the improvement predicted by simulation for service 2, 5%, is very close to the average gain of 4% for both directions obtained in the field trials.
A bigger difference can be found comparing the stops indicators. The 7% reduction predicted by the simulation becomes a 30% reduction in the field trials. This underestimate can derive from two factors:
the field trials result considers only the arterial while the simulation considers the whole network.
the calculation of stops in the NEMIS simulator is very sensitive to settings of thresholds to define a vehicle stop. The definition of a stop made by the human observer may have a certain degree of uncentainty.
As the efficiency indicators of the simualtion were very close or even
underestimating the effects of the strategy it is possible to use the results
predicted by the simulations for some other indicators that have not been
measured in the field trials. If we consider the environment indicators
that strongly depend on the stop rate and the travel time it is possible
to assume that the benefits predicted by the simulations occurred during
the field trials:
| Indicator | % change |
|---|---|
| CO Emissions | 5% |
| NOx Emissions | -2% |
| HC Emissions | -6% |
| Fuel Consumption | -3% |
Journey times for private vehicles are being reduced, as is journey time variability.
Priority is being given to public transport resulting in reduced journey times, delay and journey time variability.
ATT traffic calming using a VMS and a speed enforcement camera is succeeding in reducing the number of vehicles travelling at excessive speed and is resulting in less variability in vehicle speeds. This produces compact platoons of vehicles which are more easily controlled by the new queue management strategies.
The following parameters have been studied:
Leeds:
Click here for Figure A1.
Figure A1: Time vs Distance for a set of moving car observations
Click here for Figure A2.
Figure A2: Average normalised time vs distance for a strategy
Once a complete set of data had been collected for any of the strategies, plots were made of the average time against distance for each route covered (eg. Figure A2). This allows an indication of the performance of the strategies to be determined. On its own this only shows how the strategies have affected travel times for vehicles travelling on the selected routes. We now need to try to determine how they have changed the total travel time for all vehicles using the various parts of the network covered by the routes. Data from both routes can be combined to give average travel times along each link. Flow data from the automatic traffic counts and manual classified counts can be used to determine typical flows along each link. If the link flows are multiplied by the average link travel times and summed across all the links then the total travel time for all vehicles can be determined. Obviously checks need to be made that the flows in the network have not changed significantly between runs. This can be done by examining the ATC data collected at various points around the network. Plots were made of the daily flow profiles on the days journey time data was being collected. (eg Figures A3 and A4)
Click here for Figure A3.
Figure A3: Flows at point S13 with SCOOT+ operating
Click here for Figure A4.
Figure A4: Flows at S13 with SPOT+ operating
Statistical tests were also carried out on the flows during the two hour peak periods. At most data collection points these showed no significant differences between the flows. In those cases where there was a difference it was in the flows during the baseline data collection. As the journey time analysis has concentrated on the differences between the integrated strategies and the systems without the integrated components, these differences were not important.
AM PM Link Micro Link Micro Baseline SPOT % change Baseline SPOT % change Baseline SPOT % change Baseline SPOT % change Micro 11 Gross in 103185 97569 -5.44 361155 335222 -7.18 106041 85347 -19.52 333310 342279 2.69 Gross out 207375 164539 -20.66 177353 188122 6.07 Castel. 50595 73114 44.51 49917 68811 37.85 Micro 12 Gross in 223011 135141 -39.40 448386 224493 -49.93 237042 161249 -31.97 523511 320563 -38.77 Gross out 211263 65180 -69.15 259450 115294 -55.56 Fea 14112 24171 71.28 27020 44019 62.91 Micro13 Gross in 185483 96128 -48.17 417582 231184 -44.64 195231 128665 -34.10 485479 305781 -37.01 Gross out 194213 87068 -55.17 246008 123664 -49.73 Bibiana 37885 47989 26.67 44240 53452 20.82 Micro 14 Gross in 109199 79280 -27.40 291362 243011 -16.60 134517 72238 -46.30 396649 274885 -30.70 Gross out 165370 131333 -20.58 232506 154234 -33.66 Chiesa 16794 32397 92.91 29625 48413 63.42 Micro 15 Gross in 170389 160373 -5.88 702489 770377 9.66 116577 110985 -4.80 707643 831559 17.51 Gross out 455593 495644 8.79 486530 579307 19.07 Ala Sud 33523 50150 49.60 56004 75865 35.46 Ala Nord 42984 64210 49.38 48532 65402 34.76 Micro 16 Grosseto n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Verc. in 106404 153699 44.45 638746 737090 15.40 107284 144651 34.83 748688 817841 9.24 Botticelli 93399 108097 15.74 92504 135331 46.30 Verc.out 438944 475294 8.28 548899 537858 -2.01 Micro 17 Toscanini 44798 51891 15.83 341068 320183 -6.12 55723 59501 6.78 488736 350454 -28.29 Verc. in 150697 115422 -23.41 161650 106703 -33.99 Porpora 42405 36610 -13.67 61560 59530 -3.30 Verc. out 103168 116259 12.69 209803 124720 -40.55 Total 3200789 2861559 -10.60 3684016 3243361 -11.96
Hour SIS Code Junction 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 812 LB 180 LB 16 0.83 -18.6 -13.5 -19.1 4.82 -0.12 -1.49 -4.80 -2.59 -17.7 -6.84 -5.38 5.75 2.82 -11.3 -15.7 -13.4 850 FM 190 FM - 51.3 27.8 17.2 9.02 34.56 29.5 38.6 34.07 8.23 25.8 34.64 148 13.2 19.2 26.5 27.6 27.6 851 LB 200 LB 15 -2.53 12.7 7.24 -2.30 2.90 6 -6.22 10.95 7.05 9.54 3.75 20.9 28.2 7.58 6.03 14 22.9 870 FM 210 FM 14 -25.3 -34.0 -36.5 -22.1 2.10 -11.9 -33.7 -24.8 -23.4 -26.2 -16.6 0 -7.36 -11.6 -25 -26.6 -29.6 871 LB 220 LB 13+12 -12.2 -11.9 -3.35 -13.5 -9.36 -8.35 -5.55 -12.6 -11.9 -11.3 -14.0 -3.79 -11.5 -8.17 -9.25 -8.28 -1.64 890 FM 230 FM - -13.3 -5.37 -7.64 0 -19.6 0 -29.1 -20 -19.6 -15.8 -43.5 -35.1 -34 -38.5 -40 -39.3 -42.7 891 LB 240 LB 11 22.3 11.8 23.5 5.91 48.21 13.5 17.6 17.3 21.0 15.3 7.63 28.2 40 35.1 20.7 34.1 5.26Inbound direction
Hour SIS Code Junction 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 1644 LB 1680 LB 11 6.3 3.5 -1. -0. -8. -2. 5.7 -5. 0.8 0.4 -1. -0. 7.7 7.8 10. 9.7 12. 900 LB 1690 FM - -21 -7. -13 -20 -16 -30 -36 -33 -36 -32 -29 -22 -18 -25 -31 -34 -37 901 LB 1700 FM 12+13 -10 -5. -0. 2.6 -18 -6. -9. -6. -2. 6.9 -3. 2.7 2.8 -4. 11. 24. 23 880 FM 1710 FM - -44 -41 -38 -39 -37 -30 -45 -46 -47 -54 -44 -36 -41 -40 -42 -46 -52 881 LB 1720 LB 14+15 1.1 2.8 2.8 2.1 -3. -3. -4. -7. -12 -9. -9. -7. 2.3 1.4 7.4 5.0 2.0 860 FM 1730 FM - -23 -19 -33 -26 -32 -26 -31 -38 -38 -37 -30 -38 -40 -38 -45 -45 -44 862 LB 1740 LB 16 -15 -1. 24. 14. 43. 29. 16. 10. 8.2 11 2.0 23. 29. 44. 48. 34. 36.
ITS
home page
