Colorado Crash Data Safety Summit June 8, 2016
2015 Colorado Data* 116,388 Total Crashes 30,783 Injuries 543 Fatalities 506.015 100MVMT 88,258 Roadway Center Line Miles
*All 2015 data is preliminary
Historic Fatal and Serious Injury Trend *
* 2015 Data is preliminary
500 490 480 470 460 450 440 430 420 410 400
1.400
Fatalitiy Rate
Safety – Fatalities and Fatality Rate
1.200 1.000 0.800 0.600 0.400
VMT (100M)
Fatality Rate and VMT 1.600
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Fatalities
Why it matters: In order to achieve Colorado’s vision, Towards Zero Deaths, we need to track progress and use data to target safety improvements to maximize improvements to safety.
700
Target (2.5% reduction)
VMT (100 M)
Fatality and VMT
500
600
480
500
460
400
440
300
420 400
200
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Fatality Target (12 reduction) VMT (100 M)
* 2015 Data is preliminary
VMT (100M)
Fatality Rate
CDOT Crash Database • Contains all crashes – statewide • All agencies • PDOs, injuries and fatalities • Extracted from DOR • Coded by CDOT temp staff • Once coded/summarized, data belongs to CDOT
CDOT Crash Database Crash Coding • Geo-locating – verify crash location on non-highway crashes using: • Cross streets • Fixtures, Land marks • Cleansing the data • QA/QC • Directions • Crash type corrections • Normalizing/Uniforming • Spell check • Data consistency: Example: Colorado, Colo, CO • Identifying roadway type • Highway • Local Roads • Private Roads
Colorado Crash Record Timeline Oversight STRAC / NHTSA / FHWA
Crash Occurs Department of Transportation
State Patrol
• Investigation • Writes the report
Department of Revenue Loads and input data into database
• Download DOR Data • QA/QC • Normalize Location Coding • Create Enhanced Data
Engineering
Planning
Safety
Law Enforcement
FARS
• Investigation • Writes the report
Other Stakeholders
Crash Occurs
CDOT Data Analysis and Dissemination
Local Agencies
Legislative Departments Public Requests
Crash Occurs
Public Relations
90th Percentile
90th Percentile
90th Percentile
46 Days
30 Days
56 Days
41 Days - Paper 5 Days - Electronic
State Patrol
132 Days - Coded 90th percentile records coded - 109 days
0 9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101 105 109 113 117 121 125 129 133 137 141 145 149 153 157 161 165 169 173 177 181 186 190 194 199 203 207 211
Crashes
Days from crash date to coded date
700
600
500
400
300
200
100
0
Days
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 97 100 103 106 110 113 116 119 123 126 129 132 136 142 146 149 152 155 158 161 165
Crashes
Days between crash date and DOR load date
900
800
700
600
500
400
300
200
100
0
Days
Programs and People Using Data • Federal Highway Administration • National Highway Traffic Safety Administration • Fatality Analysis Reporting System (FARS) • Highway Safety Improvement Program • FASTER Safety Mitigation Program • Statewide Plan • Strategic Highway Safety Plan • Office of Transportation Safety • Behavioral Programs • Fatality Analysis Reporting System • Legislative Requests • Colorado State Patrol
• Colorado Department of Health and Environment • CDOT Regions • MPO/TPR’s • Public Relations • Media • Researchers • Cities • Counties • Engineers • Planners • Law Enforcement • Public • Attorneys • Marijuana Data Advisory Committee
SPF Example 160
Urban 6-Lane Freeway
140
120
APMPY
100
80
60
40
20
0 20,000
130,000 AADT Corresponds to An Expected Accident Frequency of 60 Accidents Per Mile Per Year 40,000
60,000
80,000
100,000
120,000
AADT
140,000
160,000
180,000
200,000
LOSS
Statewide Analysis
Time of Day
Day of Week
Month of Year
Contributing Factor
Roadway Description
Vision Zero Suite - Output The data from vision zero can be output to Excel. This spreadsheet can be sorted, filtered and searched.
CDOT Data • CDOT can share data with stakeholders • Data sharing = Data driven decision making • Local agencies • Law Enforcement • Planners • Systemic mitigation – Identify locations similar to locations with crashes. • Use curve and crash data to project locations before crash patterns develop.
Law Enforcement – Using Data • • • •
Single crashes are random Crash patterns indicate potential for mitigation Different from Crime Analysis Measure Effectiveness • What works and what does not?
Law Enforcement – Using Data • DUI crash pattern • Identify liquor establishments, educate • Overlay with citation data • Identify where/when enforcement is needed • Improve campaigns • Audience • Location • Dates/Times
Law Enforcement – Using Data • Staffing • Do staffing levels/shifts match needs? • Improve response time • Predict locations for minimal response times • Sunglare • Weekend tourists vs commuting traffic • Focus enforcement activities using trends • Example: Rear end vs Sideswipe • Focus on following too close or improper lane changes
Future Database Connections
CDOT Roadway Data
DOR Legal Document
CSP Citation
Health & Human Services
Weather
DOR Driver / Vehicle
CDOT Crash Data
CDPHE Death Certifica
Maintenance
Operations / INRX Hospital / EMS
Existing connection Future connection
Liquor License
LLE Citation
Judicial Citations
Others
Future of Data • Analyze systemic data errors • Provide Law Enforcement Training • Electronic Form changes • Improve efficiencies at every level • Improve location information • Improve data sharing • Improve crash report, include data needs
Future of Data • Coordination of databases • Citation • Health data • EMS • Maintenance • Roadway • Traffic Operations • Identify Over/under representation projects • Enforcement • Education • After analysis of projects • Data driven deployment of resources
Alisa Babler, PE Data Unit Manager Colorado Department of Transportation TSM&O, Traffic Data Unit
[email protected] 303-757-9967