China Safety Science Journal ›› 2017, Vol. 27 ›› Issue (5): 158-163.doi: 10.16265/j.cnki.issn1003-3033.2017.05.028

• Public Safety • Previous Articles     Next Articles

Analysis of black spot for freeway based on both statistics and hypothesis testing

MENG Xianghai, QIN Wei   

  1. School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin Heilongjiang 150090,China
  • Received:2016-12-20 Revised:2017-02-28 Online:2017-05-20 Published:2020-10-30

Abstract: In order to identify the black spots of freeway's basic segment and recognize the crash contributing factors, statistical distribution fitting and hypothesis testing were used to analyze the statistical distribution of crashes. Methods of critical crashes number determination and segment division were provided. Similarities and differences between the sliding window method and the equal-length division method were analyzed, and then probability formulas and an identification method of crash contributing factors which based on binomial test methods were given. The results showed that the crashes on freeway basic segments fit the negative binomial distribution, that the segment division method which based on negative binomial distribution is reliable, and taht although the sliding window method has a tendency to exaggerate the number of black spots, it does reflect the intensity of crashes. Case study also showed that crash contributing factors of black spots can be identified by the binomial test method.

Key words: traffic safety, freeway, black spots analysis, negative binomial distribution, binomial test, crash contributing factors

CLC Number: