Font Size: a A A

Epidemiological Analysis On Road Traffic Crash In BaNan District Of Chongqing And Application Of ARMA Model On Predictive Incidence Of General Road Traffic Crash

Posted on:2011-11-19Degree:MasterType:Thesis
Country:ChinaCandidate:L ZhangFull Text:PDF
GTID:2189360308475081Subject:Epidemiology and Health Statistics
Abstract/Summary:PDF Full Text Request
Due to Industrialization and Urbanization, road traffic safety has become a major public health challenge. In the 21st Century, road traffic crash (RTC) and road traffic injury (RTI) in China have been descended since 2002. But the huge gap between fast increase of vehicle population and slow economic development makes it hard to be optimistic about the future road safety trends in Western Region and fringe area. BaNan district is a newly developed region of Chongqing which has big area and large population. As a nearby transport hub in fringe area of Chongqing, BaNan district has become a good sample for studying the road traffic safety and effective interventions in newly developed region. Cross-sectional survey and Theoretical Epidemiology were used to study distribution features of RTC and application of prediction model. The main results are as:1.During the seven years, the general RTC was 1881 annual. The peak showed in 2004, 2102 annual. Averagely, the general RTC was 1098 per month. January, July, November, Saturday, the 1st and 14th day per month, 11 am were the highest periods of general RTC happened. The number of major RTC declined since 2002. May, September, November, Wednesday, Thursday and 10am were the highest periods of major RTC happened.2.Frontal Impact Crash was the first place crash type in both general and major RTC (24.81%, 25.04%). Both types of RTC were mainly caused by motor vehicle drivers (84.17%, 95.54%) and the highest risk comes from illegal overtaking (73.17%, 73.63%).Over 32% of general RTC were related with minibus, following by 10.1% with two-wheeled motor cycle; the sequence of major RTC was minibus (18.7%), light truck (14.3%) and two-wheeled motor cycle (14.1%). Right turn (41.35%, 21.43%) and U turn (17.55%, 15.98%) were the most risky actions.Driving experience is highly related with incidence rate of RTC. Drivers who drove less than 5 years (especially less than 1 year) caused 55.51% of general RTC and 58.74% of major RTC.The traffic control in 63.94% of general RTC crash site and 61.17% of major RTC crash site were conducted only by Road traffic signs and markings. Some crash site even had no control at all (28.44%, 32.97%). Both types of RTC mostly occurred on hybrid lane (59.28%, 70.36%), and most crash site (84.43%, 78.01%) do not have Sidewalk separation.3.ARMA model's prediction accuracy maintains about 80%, and GM (1, 1) model's prediction does not fit in analyzing the dynamic trend of general RTC data.Based on the results above, the conclusions were made as following:1.The general RTC in BaNan district increased annually, while the major RTC declined annually during the seven years. Time distribution was significant. The fundamental reason lead to RTC was the inadequate capability of vehicle drivers. The lack of Road traffic signs and effective management also played an important role.2.The ARMA model is more efficient than the GM (1, 1) model in predicting the frequency of general RTC. It has high accuracy and short-term prediction is better than long-term prediction.
Keywords/Search Tags:Fringe area, road traffic crash, epidemiology, time series, predictive model
PDF Full Text Request
Related items