{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

9_22_11_EB_state_exs

# 9_22_11_EB_state_exs - Compare Classic w Rate Control OBi...

This preview shows pages 1–9. Sign up to view the full content.

CE4001 Transportation Safety 1 Compare Classic w/ Rate Control S K XA OBi * Vi Vi XS K XS OBRi 2 1 5 . 0

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
CE4001 Transportation Safety 2 Empirical Bayesian Method Based on conditional probability, or compute posterior based on prior i.e. the probability that something is true given the knowledge that something else has occurred.
CE4001 Transportation Safety 3 Assuming use of the total entering volume for an I/S to compute an overall intersection crash rate is not logical Why not? e.g. compare 2 I/S I/S 1. high percentage of through traffic I/S 2. high percentage of left turning traffic Empirical Bayesian Method

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
CE4001 Transportation Safety 4 Assuming use of the total entering volume for an I/S to compute an overall intersection crash rate is not logical Why not? Since left turn movement might result in accidents more often, I/S 2 will be ranked higher than I/S 1 But I/S 2 might be comparable to other left turn locations Empirical Bayesian Method
CE4001 Transportation Safety 5 Bayesian method divides crashes patterns at an intersection into 15 different vehicle pair movements. For example, popular patterns for crashes: 1. Rear-end pre-stop line 2. Rear-end post-stop line 4. Right angle through’s 6. Left turn with opposing through Empirical Bayesian Method

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
CE4001 Transportation Safety 6 Empirical Bayesian Equation Coefficients (Hauer 1988, Toronto) Pattern Model Form Coefficient bo b2 1 RRpre E{m}=bo*F 0.2052x10 -6 N/A 2 RRpost E{m}=bo*F 0.1014x10 -6 N/A 4 RAth E{m}=bo*F 2 b2 8.1296x10 -6 0.3662 6 LToppth E{m}=bo*F*F 3 b2 0.0418x10 -6 0.4634
CE4001 Transportation Safety 7 Table shows E{m}, or the expected mean number of crashes per hour for each pattern F is the primary direction flow (primary through) F2 is the secondary direction flow (right angle through) F3 is the tertiary direction flow (opposing left turn) Pattern Model Form Coefficient bo b2 1 RRpre E{m}=bo*F 0.2052x10 -6 N/A 2 RRpost E{m}=bo*F 0.1014x10 -6 N/A 4 RAth E{m}=bo*F 2 b2 8.1296x10 -6 0.3662 6 LToppth E{m}=bo*F*F 3 b2 0.0418x10 -6 0.4634

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document