Status Report: Evaluation of Private Sector Data in Minneapolis Shawn Turner shawn-turner@tamu.edushawn-turner@tamu.edu, 979-845-8829 Texas Transportation.
Post on 27-Mar-2015
215 Views
Preview:
Transcript
Status Report:Evaluation of Private Sector Data
in MinneapolisShawn Turner
shawn-turner@tamu.edu, 979-845-8829Texas Transportation Institute
Mobility Measurement Pooled Fund StudyBaltimore, July 22, 2008
Average Speed Data Provided by Inrix
• 15-minute average speeds for each day of the week – from Jan 2006 through Feb 2008 (26 months)– Location– Day-of-week/time (672 bins)– Average speed– Number of data samples
• 60-minute average speeds for each day of the week and month of the year - from Jan 2006 through Feb 2008 (26 months)– Location– Day-of-week/time (168 bins) for 26 months– Average speed– Number of data samples
2
“Reliability” Statistics Provided by Inrix
• 15-minute by day-of-week statistics from speed distribution– Location– Day-of-week/time– Average speed– Speed percentiles: 10, 15, 25, 50, 85– Failure rate – frequency of time below 30, 50, 60
mph
3
Other Data Provided by Inrix
• Location referencing tables– The “decoder ring” for locations– TMC: Traffic Message Channel, a location
referencing method (de facto standard) that started in Europe and is popular among traffic information and navigation/map database companies
4
Analysis Steps
• Translate TMC linear referencing• Aggregate data to longer road segments• Visual checks of basic traffic patterns/trends• Compare data with other sources– MnDOT freeway fixed-point detectors (I-494)– MnDOT “manual floating car” signal timing runs
• Sample size analysis (Confidential)• Merge with HPMS/other DOT databases (TBD)• Calculate performance measures (TBD)
5
Case Study Corridors
6
MN 65
US 61
US 10/US 61MN 7
MN 55
US 169
US 10/US 169
CO HWY 14
I-494
HWY 13
Example Segmentation: MN 7
7
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Ave
rage
Corr
idor Spee
d (
mph)
Time of Average Weekday
US 169 NB: HWY 130 to HWY 610
Posted speed limit is 55 mph
15-Minute Average Weekday Speeds, Jan 2006 through Feb 2008
8
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Ave
rage
Corr
idor Spee
d (
mph)
Time of Average Weekday
US 169 SB: HWY 610 to HWY 130
Posted speed limit is 55 mph
15-Minute Average Weekday Speeds, Jan 2006 through Feb 2008
9
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Ave
rage
Corr
idor Spee
d (
mph)
Time of Average Weekday
US 169 NB: HWY 130 to HWY 610
2007 - 1
2007 - 2
2007 - 3
2007 - 4
2007 - 5
2007 - 6
2007 - 7
2007 - 8
2007 - 9
2007 - 10
2007 - 11
2007 - 12
Hourly Average Weekday Speeds by Month in 2007
Postedspeed limit is 55 mph
10
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Ave
rage
Corr
idor Spee
d (
mph)
Time of Average Weekday
US 169 SB: HWY 610 to HWY 130
2007 - 1
2007 - 2
2007 - 3
2007 - 4
2007 - 5
2007 - 6
2007 - 7
2007 - 8
2007 - 9
2007 - 10
2007 - 11
2007 - 12
Hourly Average Weekday Speeds by Month in 2007
Postedspeed limit is 55 mph
11
Comparison with MnDOT Fixed-Point Detectors along I-494
• Not “apples-apples” comparison• Fixed vs. mobile point speed measurement• Population vs. GPS probe sample
12
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along I-494(15-Minute Intervals, Jan 2006 through Feb 2008)
MnDOT Detectors Inrix Probes Mean Absolute Difference 13
0
100
200
300
400
500
600
-50 -40 -30 -20 -10 0 10 20 30 40 50
Fre
quency
Difference (MnDOT - Inrix)
Distribution of Average Speed Differences along I-494(15-Minute Intervals, Jan 2006 through Feb 2008)
Inrix speeds faster than MnDOT speeds Inrix speeds slower than MnDOT speeds
14
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along I-494(60-Minute Intervals, Jan-Feb 2008)
MnDOT Detectors Inrix Probes Mean Absolute Difference 15
0
20
40
60
80
100
120
-50 -40 -30 -20 -10 0 10 20 30 40 50
Fre
quency
Difference (MnDOT - Inrix)
Distribution of Average Speed Differences along I-494(60-Minute Intervals, Jan-Feb 2008)
Inrix speeds faster than MnDOT speeds Inrix speeds slower than MnDOT speeds
16
Comparison with MnDOT Test Vehicle Runs (Signal Retiming)
• Still not “apples-apples” comparison• Segment level – not link-by-link• Specific days vs. monthly average
17
3 Comparison Corridors
MN 7MN 65
MN 55
18
19
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along TH 55 EB
Hourly Average Weekday Speeds by Month in May 2007
Posted speed limit is 55 mph
Inrix Monthly Average
MnDOT Average
20
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along TH 55 WB
Hourly Average Weekday Speeds by Month in May 2007
Posted speed limit is 55 mph
Inrix Monthly Average
MnDOT Average
21
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along MN 65 NB
Hourly Average Weekday Speeds
MnDOT Average(Aug-Sep2003)
Inrix Average(Aug-Sep2006,2007)
22
0
10
20
30
40
50
60
70
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Avera
ge S
peed (
mph)
Time of Average Weekday
Comparison of Average Weekday Speeds along MN 65 SB
Hourly Average Weekday Speeds
MnDOT Average(Aug-Sep2003)
Inrix Average(Aug-Sep2006,2007)
27
0
50
100
150
200
250
300
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 15-Minute Samples in 26 MonthsAveraged across all links (both directions) on each corridor
US 10/US 169
I-494
HWY 13
US 169
MN 65
US 10/US 61
HWY 14
HWY 55
MN 7
US 61
28
0
10
20
30
40
50
60
70
80
90
100
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 15-Minute Samples in 26 MonthsAveraged across all corridors for each day of the week
Sunday Monday Tuesday Wednesday Thursday Friday Saturday
29
0
2
4
6
8
10
12
14
16
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 60-Minute Samples in January 2006Averaged across all links (both directions) on each corridor
US 10/US 169
HWY 13
US 169
MN 65
I-494
HWY 14
HWY 55
MN 7
US 61
30
0
2
4
6
8
10
12
14
16
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 60-Minute Samples in January 2008Averaged across all links (both directions) for each day of the week
US 10/US 169
HWY 13
I-494
US 10/US 61
MN 65
HWY 14
MN 7
HWY 55
US 169
US 61
31
0
2
4
6
8
10
12
14
16
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 60-Minute Samples in January 2006Averaged across all links (both directions) on each corridor
Sunday Monday Tuesday Wednesday Thursday Friday Saturday
32
0
1
2
3
4
5
6
7
12 AM 2 AM 4 AM 6 AM 8 AM 10 AM 12 PM 2 PM 4 PM 6 PM 8 PM 10 PM
Num
ber of Speed D
ata
Sam
ple
s
Time of Average Weekday
Total Number of 60-Minute Samples in January 2008Averaged across all links (both directions) for each day of the week
Sunday Monday Tuesday Wednesday Thursday Friday Saturday
Preliminary Findings-1
• Samples (actual numbers confidential)– More than you get from special studies, but
spread over numerous days– Current probes are mostly fleet, but vehicle mix
will evolve to include consumer vehicles– Samples have increased 2-3X since 2006– Samples most dense during workday, few during
overnight or weekends (but overnight/weekend not critical for congestion analysis)
33
Preliminary Findings-2
• Average Speeds– Visual checks show reasonable data during workday
hours– Inrix speeds consistently slower than MnDOT
freeway detectors• Vehicle bias – Inrix mostly commercial vehicles• Speed measurement – Mn/DOT has fixed locations
– Inrix speeds in range of arterial travel time runs– Larger month-to-month variation on arterial streets
than freeways34
Conclusions
• Applications– Segment level, maybe link (not intersection)– Historical average speeds• Hourly values on quarterly or maybe monthly• 15-minute values on at least annual basis
– Reliability statistics• Have not yet analyzed Inrix “failure” statistics• Instinct says we need more samples to represent day-
to-day variation
35
Conclusions• Possible Applications– Identifying most congested routes or segments for
further study and improvement– Ranking or prioritizing routes for funding based on
average congestion levels– Monitoring and identifying routes with the highest
increase in average congestion levels (on monthly or annual basis)
– Using as a general indicator in before-after studies of mobility improvements
– Developing performance measures on an aggregate basis.
36
Recommendations
• Proceed with more in-depth pilot/feasibility efforts – produce prototype reports– Have only taken preliminary look– Still some work needed before we get
performance measures– Private sector data will improve over time
37
Implications for Future
• There will be a private sector market for historical speeds/travel times– How can DOTs take advantage of this?– How to integrate data and address VMT?
• Different data sources– Each measurement technique has bias/error (until we have
GPS probes everywhere) – Use caution in combining/comparing data from different
measurement techniques – Adjusting for method bias?
38
top related