Advent of Big Data in Urban Transportation for Smart Cities: Current Progress, Trends, and Future Challenges

Introduction

Recently, the generation of traffic and transportation data in terms of their speed and volume have transcended at a higher scale compared to those methods that were utilised during the last few decades. The introduction of latest digital techniques like artificial intelligence, Internet of Things (IoT), data mining, social networks, the growth of smart cities (Bello-Orgaz et al., 2016) (Castellanos et al., 2021), recent improvements in wireless technologies, and the extensive usage of cost effective sensors and mobile devices has substantially upgraded humans perception on real-time traffic and transport mobility (Meekan et al., 2017) (Kaffash et al., 2021). Since Intelligent Transportation Systems (ITS) framework and execution is based on traffic mobility data which in terms of volume, diversity (source and format), and its variable nature requires data intensive techniques like querying and analysis, integration, high performance computing, visualisation of extensive real-time systems are necessary (Khattak, 2017; Andrienko et al., 2017; Amini et al., 2017; Usman et al., 2020). Traffic congestion and road accidents are the biggest challenges faced in urban cities that requires immediate attention (Matcha et al., 2020; Ng, 2021). Hence, a more distinctive and enhanced methods of transport data collection, transmission, storage, fusion, extraction and processing are necessary (Matcha et al., 2022; Stathopoulos et al., 2017). However, it is widely admitted that the present ITS implementation is limited to data monitoring and evaluation applications (Suh et al., 2017). The above-mentioned challenges were not fulfilled by current ITS applications for efficient monitoring, decision-making and realistic data management applications. Moreover, the transport facilities, projects and operations can be improved by implementation of advanced aggregation, integration and progressive learning approaches which is possible by inflow of massive amount of real-time data streaming sources for a better insight and policy making decisions. The classification of Big Data is shown in Figure 1.

In this scenario, Big Data has been suggested as an essential technology for the transportation sector (Rusitschka & Curry, 2016; Arfat et al., 2020b). This technological exemplar can adapt, administer, and analyse huge amounts of organised and unorganised data by providing solutions and tools to upgrade the transport sector and face future challenges. The solution requires developing state-of-the-art ITS and transportability services by extricating the worth and knowledge of the available bulk data established on the concepts and techniques of the Big Data. Finally, it is to make sure that the transport sector evolves to present a sustainable and well-developed transport services and experiences to its users by deriving value from its data. The harvesting and mining of data required to predict the patterns in transport mobility behaviour to improve safety and security requires predictive analytics, which is a technological equivalent to Big Data. For instance, infrastructure monitoring is an application based on Big Data in Transportation (Jeong et al., 2016), establishment of transport added-valued services (Mehmood & Graham, 2015; Fonzone et al., 2016; Su et al., 2016), better interpretation of road users’ requirements (Chen et al., 2016b) and foreseeing the evolution of traffic flow in urban areas (Kitchin, 2014).

Figure 1.

Big Data Classification