1. Robert Herling, Planner Duluth-Superior Metropolitan Interstate Council rherling@ardc.org(218)–529-7573 A Quantitative Assessment Method For System Safety & Efficiency September 18, 2008

2. NOT scientific NOT statistically significant IS a way to measure relative performance IS inexpensive IS fair & balanced The methodology is…

3. * * Transportation in Duluth- Superior

4. Goal of Transportation Systems Management (TSM): Improved safety & efficiency through low-cost, high-benefit solutions Improved safety & efficiency through low-cost, high-benefit solutions perform a system-wide assessment perform a system-wide assessment prioritize locations by needs prioritize locations by needs provide a method of monitoring provide a method of monitoring MIC Project Objectives

5. 3 crashes per year Reduced crash rate High severity rate 3 Crashes per year Moderate crash rate Low severity rate YEAR 1 6 Crashes per year High crash rate High severity rate Strategy: identify, quantify, and monitor… YEAR 3 YEAR 5

6. running a system-wide analysis with any frequency balancing safety and efficiency needs balancing spot locations and road segments comparing needs between two states Challenges: Solutions: thresholds to limit number of locations analyzed address spot locations and road segments separately do two separate assessments (MN & WI)

7. Started with a previous concept: Adjusted Scale 5 year crash trend Current crash severity Level of service Functional class 123456 0246810 ABCDEF Major/ Major Major/ Minor Major/ Collector Minor/ Minor Minor/ Collector Collector/ Collector

8. Only analyzed spot locations Safety threshold: 3 or more crashes in most recent year Mobility threshold: LOS D or poorer Did not compare MN results with WI results Initial attempt in 2005:

9. First assessment done in 2005: Locations ranked with weight values (LOS given more influence) For each location, analysis of conditions contributing to poor safety & capacity

10. Too simplified Crash rate not the best measure Too much emphasis put on LOS Problems with 2005 assessment: Suggested improvements: Use similar measures for both capacity and safety Use same number of measures for both Get rid of the weights

11. Crash frequency (Quantity) Crash severity rate (Severity) Average increase in crash rate (Change) Safety performance measures: Mobility performance measures: AADT per lane mile (Quantity) LOS (Severity) % change in AADT (Change) Improved methodology:

12. (crashes 2002 + 2003 + 2004 + 2005 + 2006 ) / 4 Average number of crashes per year Quantity measure for safety:

13. Number of equivalent property damage crashes per million vehicles Severity measure for safety: total crashes x property damage crashes ( x 1x 1 ) injury crashes ( x 2x 2 ) fatality crashes ( x 3x 3 ) ++ [] total years x 365 days x AADT

14. 0.5 0 1 year 0.5 0 Change in Rate 5 yr Avg. 2002 to 2003 State Hwy 194 & Mesaba Ave E 4 th St & 6 th Ave E 2002 to 2003 [( 2003 + 2004 + 2005 + 2006 ) / 4 ] – 2003 Average change in crash rate vs. base years change in crash rate (2002 to 2003) Change measure for safety: -0.5 - 0.34 + 0.20 Change in Rate 1 year

15. AADT per lane (factored by traffic control type) Quantity measure for mobility: AADT 2006 / number of lanes () x traffic control device factor where: No control (not an intersection) =.01 Roundabout =.02 Signal (protected L-turn for all) =.03 Signal (protected L-turn for some) =.04 Signal (no protected L-turns) =.05 Stop or Yield =.06 1 2 3 4 5 6 7 8 910

16. LOS as identified by TP Plus model Severity measure for mobility: A, B, C, D, E, F () LOS A = 1 LOS B = 2 LOS C = 3 LOS E = 5 LOS F = 6 LOS A = 4 where:

17. % change in AADT from 1 st year to last year (adjusted by func. class) Change measure for mobility: ( AADT 2006 - AADT 2002 ) / AADT 2002 ][ (FC 1 + FC 2 + … FC n) / 100 + where: Major arterial = 5 Minor arterial = 4 Major collector = 3 Major collector = 2 Local road = 1

18. Indexing How can these things be compared? Composite scoring I j = -100(X/R)+100 n C = Σ(I j ) j - 1 Creates an individual score between 0 - 100 Combines all individual scores for a final score between 0 - 600

19. Indexing I j = -100(X/R)+100 /-100 x [ worst severity rate ( ] best severity rate ) _ +100 worst severity rate )( _ locationsSeverityrate example:

20. Composite scoring n C = Σ(I j ) j - 1 + index value for % change In crash rate ( ) index value for crash frequency ( ) index value for severity rate ( ) + index value for LOS ( ) index value for AADT per lane mi ( ) index value for % change in AADT ( ) +++

21. Can prioritize locations by composite score INTERSECTIONS COMPOSITE SCORE

22. Or analyze locations in how they rank by safety or efficiency INTERSECTIONS COMPOSITE SCORE

23. E 4 th St - Heading NE 6 th Ave E - Heading SW E 4 th St & 6 th Ave E: Mobility Issues E 4 th St 6 th Ave E LOS D LOS C * Lack of protected L-turns on E 4 th St

24. W 7 th St Skyline Pkwy W 7 th St - Heading SW Skyline Pkwy - Heading NE Skyline Pkwy & W 7 th St: Safety Issues * Sight distance issues

25. NOT scientific NOT statistically significant IS a way to measure relative performance IS inexpensive IS fair & balanced The methodology is…

26. Questions? Robert Herling, Planner: rherling@ardc.org Duluth-Superior Metropolitan Interstate Council (MIC): www.dsmic.org