A Case for Machine Learning in Edge-Oriented Computing to Enhance Mobility as a Service

The study of human mobility tries to understand human flows and synergies with the geographical environment. Mobility as a Service (MaaS) is a new mobility concept that promises to revolutionize commuting by merging public and private transport providers around a common platform that travelers will use as a service, thus providing new research opportunities for human mobility. One of MaaS main offerings is the ability to calculate both routes and commuting strategies based on the availability of transports and specific user constraints. In this work, we discuss how Edge-Oriented Computing (EOC) and Machine Learning (ML) can contribute to extending the reach of MaaS in the upcoming years. EOC enables technologies to perform computation at the edge of the network, reducing latency and communication overheads, which 5G technologies are committed to further diminish, thus benefitting the proliferation of MaaS. Also, ML techniques are one of the most robust approaches for planning routes and predicting future movements. Finally, we present open research topics that will promote the attractiveness of MaaS.