Determining Factors Affecting Acceptance of Autonomous Vehicles using Statistical and Machine Learning Models

The aim of this study was to find out the risks and perceptions related to the acceptance of Autonomous Vehicles (AVs) with regards to different aspects of society. An online survey was used for collection of stated preference data. The data of 465 respondents was deemed suitable for the analysis of this study. Comparison with traditional vehicles and willingness to use had the highest ratings while being tech-savvy had the lowest ratings. Parametric analysis and prediction model were used to analyze the relationships between the willingness to use and participants’ characteristics and opinions. The model was developed using artificial neural network. The results show that gender, age, affinity for technology and comparison with traditional vehicles seem to have a significant impact on the perception of participants. This was shown by the parametric analysis performed at a significance level of 5% and later confirmed by the model. The model showed the highest importance of being tech-savvy with 0.76 index followed by comparison with an index of 0.74. A comparison with a similar study from Saudi Arabia shows that drivers in these countries have a significantly different perception related to AVs. Highlights: • Determination of acceptance of Autonomous Vehicles (Avs) using various factors. • Statistical analysis and artificial neural networks used. • Results compared with a neighboring country and previous studies. • Providing recommendations to promote AVs and realize their full potential.