Development of 3W-LOS Prediction Model for Urban Roadways using Adaptive Neuro Fuzzy Interface System

In order to transform cities into more liveable, safe, and sustainable places, we must shift our mobility paradigms. As one auspicious concept amongst urban transportation facilities, para transit modes facilitate urban transportation into small, efficient, and affordable vehicles that are flexibly operated on any infrastructure i.e. specifically motorized three-wheelers being the most prominent subject vehicle of para-transit mode in India, yielding the potential to make public transit more convenient, affordable, and sustainable all at once. Considering this, service quality prediction for motorised three-wheelers at urban roadways is analysed involving the data collected from 50 road segments from 5 cities. Adaptive Neuro Fuzzy Interface System (ANFIS) is used for the development of Three-Wheeler Level of Service (3W-LOS) prediction model, which simplifies the complex and wide input space. For the above quantitative parameters as road geometric features, flow parameters of traffic, built environment factors are considered and also for predicting the service level qualitative survey is done. Information of the survey participant (age, sex, driving experience, type of vehicle, education qualification, frequency of travel, etc.). A set of attributes are shown to drivers about road segment LOS and are asked to rate the facility on a Likert scale of 1-6 where, ‘1’ indicates ‘highly dissatisfied’ and ‘6’ indicates ‘highly satisfied’. Overall satisfaction score (OS) of road links, taking the average of driver's rating is obtained. The coefficient of determination for training and testing is found to be 0.8809 and 0.8239 respectively by using the ANFIS output for 3W-LOS prediction over Indian roadway segments.