INVESTIGATING INFLUENCE OF PAVEMENT CONDITION ON ROAD ACCIDENTS (Case Study: National Road Network in Sleman, D.I. Yogyakarta)
R.M. ALDIREZA FIRMANSYAH, Dr. Ir. Latif Budi Suparma, M.Sc.
2013 | Tesis | S2 Mag. S. & T.Transportasi-
There are three major factors; the driver, the vehicles, and the environment such as the road condition that usually caused a road accident. A combination of those factors are often happens as well. Drivers errors are usually occur not only because of their mistakes, but also because the road condition or vehicles performance are affecting them. Concentrating too much effort on the human factor as the source of an accident would conceal the ability to see other opportunity that contributes to accident rates, such as the road condition. Currently the research on effect of pavement condition on road accident is limited, especially in Indonesia. Thus this thesis is aim to assess the pavement conditions and road accident on National Road Network, particularly in DI Yogyakarta Province, to predict the relationship of pavement condition with road accident by construct statistical model and to investigate the significant accident condition and road user that relates to pavement conditions; The dissertation is studying the effect of road pavement condition on road accidents in DI Yogyakarta Province national road network. Road links that selected on the study are interurban road with 2 lanes with divided median. The analysis was carried out by collecting data pavement condition and police’s accident reports. The negative binomial regression is uses in this study in order to estimate the effect of road pavement condition on road accident. Several accident data such as total accident, accident in daylight, in dark, the severity of accident as the fatality are used as the dependent variable. While the International Roughness Index and the Surface Distress Index was used as the independent variable. SPSS software were used for develop the regression model. In general the road pavement condition on the links that observed in this thesis is considerably in good condition. On overall link 015, 019, and 019-11K on National Road Network in DI Yogyakarta is 56.41% is in good condition, while 34.36% are in medium condition. The modelling develop in this study show that SDI and IRI value are not good regressors as the model failed the significant test. This might happen because the link that have had selected are considerably short. The most venerable road user according to the accident data as for both multiple and accident crash is motorcyclist.
Kata Kunci : road accident, regression, pavement deterioration
