Thesis Archive
A predictive model of motorcycle accident occurence considering driver and environment dependent factors on main roads in Quezon City (2010)
Glaiza Marie T. Flores
Maria Patricia T. Gotohio
Noel Gabriel C. Paras
Abstract:
-In 2009, data suggested that more than 700,000 motorcycles including tricycles newly registered and almost two million renewed their registration in the Land Transportation Office (LTO) in National Capital Region (NCR). Motorcycles are becoming increasingly prevalent, as they are the one if not the most economic mode of transportation. However, according to the Metro Manila Development Authority, the trend of motorcycle accidents from 2005 to 2009 has been constantly increasing. Since motorcycle accidents imply, property damages, injuries and mortality, there is a need to identify factors that could contribute to accident occurrence.
It is also found that there is a lack of exploratory models on how contextual and individual factors separately and interactively contribute to injury causation. Furthermore, studies on motorcycle accidents in the Philippines are very limited.
With this, based on previous studies and data of motorcycle accidents from the Quezon City Police District, motorcycle accidents could be attributed to a number of environment and driver-related factors. The factors identified were age, driving behavior, movement, road character, day and time of the accident, lighting condition, surface condition, weather condition and junction type.
The study has formulated a survey in order to gather data on the said factors with respect to accident occurrence. Two hundred thirty-six (236) survey respondents consist of motorcycle drivers who have been renewing their licenses in LTO-Quezon City from March to June 2010. Eighty percent of the data is used in developing the predictive model and 20% is for validating it. Binary logistic regression is utilized to analyze, which factors would contribute to the probability of an accident. Results showed that weather, driver behavior and junction type could significantly affect motorcycle accidents. Reducing the model further, weather and the interaction of driver behavior and age remained significant.
A series of policy revisions and programs are proposed based on the results of the study. Recommendations are made for further research. A series of policy revisions and programs are proposed based on the results of the study. It could be derived from the study that education and awareness could lessen the occurrence of accident, which can be achieved using mass media like television advisories, the Internet, print media. This could also be achieved by implementing a Compulsory Basic Training for people who wants to drive motorcycles. In this way, it can be ensured that one acquires the necessary knowledge of motorcycling and had demonstrated the basic practical skills and attitudes to make him/her safe on the road.
It is also important to ensure road signs and warnings are clear and high-risk road sections are to be well lighted. This could be done by introducing a road safety audit, wherein proposed or current road designs are evaluated to identify potential safety hazards, which may affect any type of road user and suggest measures to eliminate such hazards.
Moreover, recommendations are made for further research. Since the one of the concerns in the study is the reliability of the data, an improved data collection is suggested. The availability of more reliable data can pave the way for better studies. Other environment and driver dependent variables can also be explored to investigate other causal relationship that can affect motorcycle accidents. The feasibility of decreasing speed limits, adding motorcycle lanes, and implementing Intersection Decision Support System can also be conducted to determine if it can lessen the occurrence of motorcycle accidents.
Adviser:
Mr. Rosemary Seva