Big data revolution for traffic congestion
5th Mar, 2017
stewiek
5th Mar, 2017
stewiek

Dr Rahat Iqbal, an expert in Big Data Analytics, recently paid a productive visit to the Governor office of Jakarta, Indonesia to discuss potential solutions for traffic congestion problems. Dr Iqbal is a Reader in Human-Centred Technology in the School of Computing, Electronics and Mathematics at Coventry University, and a CEO of Interactive Coventry Ltd, a supplier of high-tech intelligent data-driven solutions. Mrs Barbara Howell, Associate Dean (International) also attended the meeting together with Ms Yanti Amran, a local part-time Consultant for the University Jakarta office. The visit was backed by Mr Paul Fairburn (Director of Enterprise and Innovation); Professor Mike Hardy (Executive Director, Centre for Trust, Peace and Social Relations and Professor Damien Foster (Head of the School). Present from the Governor’s office were Mr Setiaji (Head of Jakarta Smart City Unit), Mr Diory Paulus Damanik (Head of Data & Analytics), JSC team and many other officials from the Government Transport Wing where Dr. Iqbal presented the state of the art in machine learning approaches relating to the prediction and management of traffic congestion. Mrs Howell supported Dr Iqbal by briefing the officials with the example of the London traffic management system.

Fig. 1 Prototype application (heat map of traffic flow for Jakarta city visualization window)

Dr Iqbal suggested novel solutions for managing traffic congestion which were based on Big Data Analytics. During the subsequent discussion, prototype systems (as shown above) applied to a real-world dataset were demonstrated. The screenshot (Fig 1.) shows a heat map indicating the volume of traffic flow for Jakarta city. The wide range of proposed solutions include traffic visualisation, prediction and optimisation. Additionally, a novel solution based on multi-agent technology was suggested which could analyse the potential impact of policies relating to transport. The government officials showed a strong interest in these projects and a willingness to continue this collaboration with Coventry University which could lead to a mutually beneficial strategic partnership. The suggested solutions were based on technology patented by Interactive Coventry Ltd using deep machine learning techniques. The patent introduces a novel biologically inspired universal generative modelling approach called a Hierarchical Spatial-Temporal State Machine. The patented approach was developed on the understanding of the brain, its structure and functionality. This technique, is capable of modelling and predicting complex spatial-temporal patterns in data from which it can predict the future states of a system, based on its previous behaviour, while taking into account significant noise in the data. The approach can learn automatically the complex patterns in traffic data in order to identify and predict traffic congestion. This gives it a competitive advantage over rival methods where substantive human supervision is required. Due to its unique capability for handling data invariances, the method is able to handle a broad range of data types to discover patterns which are too complex to be identified by humans or standard machine learning techniques.