Abstract
Enhancing route planning the ability to improve route planning as applied to MaaS is paramount to improving mobility in urban areas and the experience of it. This work focuses on deep learning methods for improving route planning in context of MaaS. suggest the deep learning model employing traffic data, the user preference and real-time environmental data to develop efficient routes planning. The model uses convolutional neural network (CNN) for spatial-temporal data analysis extracting features related to graphs and exploits the recurrent neural network (RNN) for sequential data analysis such as traffic flow and time series. Thus, by using this deep learning mechanism the system is capable of updating routes whenever there is a change in the environment thus offering persons efficient and customarily satisfying routes.