Study Details
Study Title: Using multivariate adaptive regression splines (MARS) to develop crash modification factors for urban freeway interchange influence areas
Authors: Haleem et al.
Publication Date:JUN, 2013
Abstract: Crash modification factors (CMFs) are used to measure the safety impacts of changes in specific geometric characteristics. Their development has gained much interest following the adoption of CMFs by the recently released Highway Safety Manual (HSM) and SafetyAnalyst tool in the United States. This paper describes a study to develop CMFs for interchange influence areas on urban freeways in the state of Florida. Despite the very different traffic and geometric conditions that exist in interchange influence areas, most previous studies have not separated them from the rest of the freeway system in their analyses. In this study, a promising data mining method known as multivariate adaptive regression splines (MARS) was applied to develop CMFs for median width and inside and outside shoulder widths for "total" and "fatal and injury" (FI) crashes. In addition, CMFs were also developed for the two most frequent crash types, i.e., rear-end and sideswipe. MARS is characterized by its ability to accommodate the nonlinearity in crash predictors and to allow the impact of more than one geometric variable to be simultaneously considered. The methodology further implements crash predictions from the model to identify changes in geometric design features. Four years of crashes from 2007 to 2010 were used in the analysis and the results showed that MARS's prediction capability and goodness-of-fit statistics outperformed those of the negative binomial model. The influential variables identified included the outside and inside shoulder widths, median width, lane width, traffic volume, and shoulder type. It was deduced that a 2-ft increase in the outside and inside shoulders (from 10 ft to 12 ft) reduces FI crashes by 10% and 33%, respectively. Further, a 42-ft reduction in the median width (from 64 ft to 22 ft) increases the rear-end, total, and FI crashes by 473%, 263%, and 223%, respectively.
Study Citation: Haleem, K, A. Gan, and J. Lu. "Using multivariate adaptive regression splines (MARS) to develop crash modification factors for urban freeway interchange influence areas". Accident Analysis and Prevention, Vol. 55, (2013) pp. 12-21.
Related Citations: Haleem, K., A. Gan, and J. Lu, "Developing Crash Modification Factors for Interchange Influence Areas on Urban Four-Lane Freeways Using Multivariate Adaptive Regression Splines." TRB 91st Annual Meeting Compendium of Papers, Washington, D.C, (2012).
Comments: Study 276 (TRB Paper) and Study 351 (AAP Paper) are from the same authors and present different results. The differences in the CMFs between the two studies could be due to the different time periods analyzed (Study 276 used 2005 – 2008 FL data; Study 351 uses 2007 – 2010 FL data), size of the dataset (the dataset used in Study 276 has 12,078 total crashes – both calibration and prediction dataset, combined; the dataset used in Study 351 has 58,394 total crashes – both calibration and prediction dataset, combined) and base conditions for inside and outside shoulder widths (Study 276 used 4 ft inside and 10 ft outside shoulder width as the base condition; Study 351 used 10 ft inside and 10 ft outside shoulder width as the base condition).
Both studies show a decrease in crashes, in either direction of the base condition. The difference in the base condition coupled with unknown disaggregate sample sizes makes it harder to interpret these results. It is expected that the crashes would decrease with an increase in shoulder width; however, the authors state that reducing the shoulder width from the base condition would lead to most drivers being unable to park alongside the narrow width and could easily change one lane (along a two-lane per direction freeway) to park on the right (outside) shoulder width. This reduces the risk of fatal crashes with vehicles traveling in the higher-speed leftmost lane due to stopping on the left (inside) shoulder. FHWA encourages customers to review the full study to determine the applicability of the CMFs.
CMFs Associated With This Study
Category: Access management
Countermeasure: Decrease median width from 64ft to 22ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
3.629 | -262.9 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
3.227 | -222.7 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
5.732 | -473.2 | | Rear end | All | Principal Arterial Other Freeways and Expressways | Urban |
4.184 | -318.4 | | Sideswipe | All | Principal Arterial Other Freeways and Expressways | Urban |
Countermeasure: Decrease median width from 64ft to 40ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
1.071 | -7.1 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
1.073 | -7.3 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
1.43 | -43 | | Rear end | All | Principal Arterial Other Freeways and Expressways | Urban |
1.151 | -15.1 | | Sideswipe | All | Principal Arterial Other Freeways and Expressways | Urban |
Category:Shoulder treatments
Countermeasure: Decrease inside shoulder width from 10ft to 4ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
0.446 | 55.4 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
Countermeasure: Decrease paved outside shoulder width from 10ft to 8ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
1.081 | -8.1 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
1.172 | -17.2 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
1.111 | -11.1 | | Rear end | All | Principal Arterial Other Freeways and Expressways | Urban |
Countermeasure: Increase inside shoulder width from 10ft to 11ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
0.98 | 2 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
Countermeasure: Increase inside shoulder width from 10ft to 12ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
0.669 | 33.1 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
Countermeasure: Increase paved outside shoulder width from 10ft to 12ft
CMF | CRF(%) | Quality | Crash Type | Crash Severity | Roadway Type | Area Type |
---|
0.774 | 22.6 | | All | All | Principal Arterial Other Freeways and Expressways | Urban |
0.9 | 10 | | All | K,A,B,C | Principal Arterial Other Freeways and Expressways | Urban |
0.816 | 18.4 | | Rear end | All | Principal Arterial Other Freeways and Expressways | Urban |