Car-motorcycle accidents have been reported higher in recent years in Hungary due to increasing number of motorbikes on road. Car-motorcycle collisions mostly lead to fatal and seriously injured accidents mainly due to the vulnerability of motorcyclists and other related factors. The crash investigation studies aim to analyze the main contributing factors that cause fatal road accidents and injury outcomes. The main goal of this study is to evaluate and compare the contributing factors to car-motorcycle accidents in Budapest city by using a microsimulation tool. The procedure utilized the statistical analysis and data sampling to categorize car-motorcycle accidents by dominant accident types based on collision configurations. The police report is used as a data source for designated accidents and simulation models are plotted according to scale (M 1:200). The simulation crash study results observed the main contributing factors to car-motorcycle accidents such as driver behavior, rider behavior and view obstruction. The comprehensive in-depth investigation also found that most of the car drivers and riders could not perform collision avoidance manoeuvres before the collision. This study can help the traffic safety authorities to solve road safety issues by considering the main contributing factors to car-motorcycle collisions. The study also proposes safety measures to avoid car-motorcycle accidents in future.
Cotton fibre maturity is the measure of cotton’s secondary cell wall thickness. Both immature and over-mature fibres are undesirable in textile industry due to the various problems caused during different manufacturing processes. The determination of cotton fibre maturity is of vital importance and various methods and techniques have been devised to measure or calculate it. Artificial neural networks have the power to model the complex relationships between the input and output variables. Therefore, a model was developed for the prediction of cotton fibre maturity using the fibre characteristics. The results of predictive modelling showed that mean absolute error of 0.0491 was observed between the actual and predicted values, which show a high degree of accuracy for neural network modelling. Moreover, the importance of input variables was also defined.