01476155

Component

This paper evaluates how different roadway cross-sectional elements and traffic characteristics impact crashes involving transit buses and general traffic crashes. The contributing level of each of the elements is compared between the two crash scenarios. One of the elements included in the study is the impact of bus trips per day which has not been covered by many previous studies. Zero Inflation Negative Binomial (ZINB) and the standard Negative Binomial (NB) models were applied in the analysis. The ZINB method was applied to the transit bus crashes modeling due to the presence of many roadway segments with zero crash experiences. As expected, the model results presented a positive coefficient for the number of bus trips which meant that the higher the number of trips per day, the more crashes for that specific roadway segment. Other positive coefficient variables in the bus crashes model include the presence of raised medians, presence of two way left turn lanes (TWLTL), and high posted speed limits. Negative coefficient variables were found to be number of lanes, median width, lane width and shoulder width and the presence of curb and gutter. While most of the variable coefficient signs in ZINB were the same as those in standard NB, their magnitudes and significance varied across these two models. Study findings present safety factors to consider when assigning transit bus routes and scheduling on top of other general traffic safety measures.

This paper was sponsored by TRB committee AP050 Bus Transit Systems.

01470560

13-0109

English

13p

2013

Transportation Research Board 92nd Annual Meeting

Digital/other

References; Tables

Binomial distributions; Bus crashes; Bus routes; Crash analysis; Traffic characteristics; Transit buses

Zero-Inflated Negative Binomial

Tennessee

Highways; Public Transportation; Safety and Human Factors; I80: Accident Studies

Transportation Research Board Annual Meeting 2013 Paper #13-0109

TRIS, TRB, ATRI

Feb 5 2013 12:11PM