Determination Of The Risk Factors And Injury Probability For Small Car Crashes In Klang Valley, Malaysia
Sisilia, Vivi (2007) Determination Of The Risk Factors And Injury Probability For Small Car Crashes In Klang Valley, Malaysia. Masters thesis, Universiti Putra Malaysia.
In year 2004, there was a study conducted for a report to the Ministry of Transport of Malaysia indicated that 40% of motorcyclists had access to motorcars as an alternative mode of transport. As a major mode of transport in Malaysia, motorcycle revealed about 48% of the registered vehicles in this country. The statistic compilation showed that number of motorcycles registered were increase from 3,564,756 in year 1995 to 6,164,953 in year 2003. Annual accident record showed that the percentage of motorcycles involved in road accidents recorded also increased by 10.3% from 86,834 in year 2002 to 95,545 in year 2003. In comparison, as a second major mode, the number of registered motorcars also experienced a high increase from 2,532,396 to 5,428,774 within the same period. And for the accident record, in the year 2002 alone, a total of 320,719 motorcars were involved in road accidents, representing an increase of 9.7%. At the end of year 2003, the total reached 351,832.The growth of small car ownership in Malaysia has tremendously increased along with the tendency of changing the mode of transport from motorcycles to the safer type of vehicles. In the year 2003, 123,064 units of cars with engine capacity below 1000 cc were sold; 74,088 units for engine capacity 1301 cc – 1500 cc; 115,468 units for engine capacity 1501 cc – 2000 cc; 17,621 units for engine capacity 2001 cc – 2500 cc; and 6,884 units for cars with engine capacity 2501 cc – 3000 cc (JPJ, 2005). Along with the tremendous growth of vehicle ownership, especially motorcars, in Malaysia, the number of motorcars involved in road accidents has also alarmingly increased. However, the accidents involving motorcars continue to be a problem as the number has kept dramatically increasing in the last decade compared to the number of motorcycles involved in road accidents. The study of risk factors and developing a prediction model on small car crashes were needed to address several matters such as: a) tremendous increase of registered motorcars which amplifying the rates of motorcar crashes; b) a need for understanding a small car crash patterns in order to reduce the severity of its occupants and highway looses; c) currently, different countries have different policies in categorizing the small car for example by fuel, engine capacity and weight, thus there is a need for a point of reference of small car specification and categories especially that small cars reveal as the highest number of cars travelling on the road in this country. To achieve the aforementioned aim, the objective of this study was established to determine what the risk factors for small car crashes in Malaysia are and to develop a statistical model based on Binary Logistic Regression for predicting the probability of injury and non-injury accidents involving small cars.Fourteen hypotheses have been summarised from literature reviews and carried out into univariate and multivariate analyses to identify the risk factors and injury probability prediction model for small car crashes. Odds ratios have been using to determine the strength of association between the accident type and the risk factors. The accident type which was the dependent variable was categorised into two injury and non-injury accident. From the univariate analysis, the odds ratios determined that there were five risk factors of small car crashes that significantly caused injury to its drivers. Those risk factors were length of small car wheelbase, gender of small car driver, the utilization of seat belts, the changing of velocity after the impact, the extent of the damage, and the weight of the drivers. From the model development, all fourteen parameters tested using Binary Logistic Regression and the parameters sorted out into three significant small car crash predictors that causing higher probability of experiencing injury to the victims when exposed to accident. Those significant predictors were weight of the drivers, velocity changes and damage extent (χ²=2.330, df=5, p<0.802). The model indicated that injury accident was more likely for small car which driver weighed 64kg and above, where the change of velocity 17km/h and above, and that where the extent of the damage reached extent scaled-2 and above
Repository Staff Only: Edit item detail