Real time analytics of urban congestion trajectories on hadoop-mongoDB cloud ecosystem

Abstract

Cities should invest significant budgets on technologies of the Internet of things in the next few years. One of the great challenges is to align technologies related to connected objects exploring the urban-cloud computing and big data analytics to intelligent urban traffic and congestion management. Traffic congestion in urban networks takes polymorphic forms and magnitudes. Technological advances on big data analytics, distributed processing on the cloud as well as new technologies of location-based services and traffic measurement are great support for the congestion management processes. This paper mainly discussed to formulate a new model of urban traffic congestion based on the trajectories meta-model and on some new algebraic structures. The idea is to consider congestion as a space-time event and its extent as a wave spreading modelled by a space-time path trajectory. It is also considering the issue of congestion such as infections that spread in a network. We define two algebra structures that meet the expectations of congestion management. These algebraic structures generate very useful congestion patterns for network traffic control and monitoring functions. For congestion trajectories temporal queries, mining purpose, and spatio-temporal visualization, these spacetime congestion paths will be stored in a MongoDB data warehouse operating in a Hadoop cloud-based eco-system. This approach is of great simplicity we have to operate later in the traffic management schemes.