Data Fusion for MaaS: Opportunities and Challenges

Abstract

Abstract-Computer Supported Cooperative Work (CSCW)in design is an essential facilitator for the development and implementation of smart cities, where modern cooperative transportation and integrated mobility are highly demanded. Owing to greater availability of different data sources, data fusion problem in intelligent transportation systems (ITS) has been very challenging, where machine learning modelling and approaches are promising to offer an important yet comprehensive solution. In this paper, we provide an overview of the recent advances in data fusion for Mobility as a Service (MaaS), including the basics of data fusion theory and the related machine learning methods. We also highlight the opportunities and challenges on MaaS, and discuss potential future directions of research on the integrated mobility modelling.

Heuristic-Based Journey Planner for Mobility as a Service (MaaS)

“…Technology is expected to play an important role when the maturity of MaaS schemes reaches a level where multiple MaaS operators and mobility service provides need to collaborate in a single city. Data fusion [15], journey planning and ticketing applications [16] and integrated multimodal information platforms [17] are some of the technologies that need to be realised as enablers of MaaS. The above-named technologies and systems are present in most conceptual and prototype MaaS architectures proposed in the literature [10].…”

Data, AI and governance in MaaS – Leading to sustainable mobility?

“…MaaS cannot exist without data, and specifically, without data collected from users, public and private mobility providers and infrastructure (e.g. sensors, traffic lights) (Wu et al, 2018). As such, MaaS requires sufficient initial users as well as mobility providers to generate data to be operationally and commercially viable.…”

“…Its promises to transcend human cognitive capacities date back to the 1950 s yet failed to be realised by the 1980 s. Recently, AI has been revitalised in the form of machine learning (ML), particularly its subset deep learning and data mining algorithms. ML algorithms in MaaS are trained with (big) data to accomplish previously difficult and time-consuming cognitive tasks, such as predicting, modelling and clustering demand and supply (Bhavsar et al, 2017;Wu et al, 2018). Recommender systems, another type of ML, can also be used in MaaS for nudging users towards using specific modes and for data integration (Arnaoutaki et al, 2021).…”

“…The central focus of this literature is on the current and prospective applications of AI algorithms in business efficiency, data analytics and mobility management (see Leung et al, 2019;Wan & Hwang, 2018;Yigitcanlar et al, 2020). This technical literature suggests that AI can contribute to sustainable mobility by reducing traffic congestion and emissions and providing affordable/accessible transport (Nikitas et al, 2020;Wu et al, 2018). The literature that specifically links MaaS and AI covers the potential use of AI in MaaS in the following three key stages.…”

“…Hybrid governance in MaaS and what it entails for sustainable mobility remain largely understudied both in the technical and the social science literature. The role of data and AI is covered mostly by technical literature, which is typically limited to trusted and secure transactions, data privacy, standardisation, and user customisation (Abduljabbar et al, 2019;Diran et al, 2021;Nikitas et al, 2020;Paiva et al, 2021;Wu et al, 2018). Literature on MaaS governance is largely from the social sciences and focuses on new business models that emerge through MaaS and the role of multiple human actors (Audouin & Finger, 2018;Pangbourne et al, 2019;.…”