The idea that including more information in more dynamic and iterative ways is central to the promise of the big data paradigm. The hope is that via new data sources, such as remote sensors and mobile phones, the reliance on heavily simplified generalised functions for model inputs will be erased. This trade between idealised and actual empirical data will be matched with dynamic models which consider complexity at a fundamental level, inherently mirroring the systems they are attempting to replicate. Cloud computing brings the possibility of doing all of this, in less time than the simplified macro models of the past, thus enabling better answers and at the time of critical decision making junctures.

This research was task driven - the question of high speed rail versus aviation led to an investigation into the simplifications and assumptions that back up many of the commonly held beliefs on the sustainability of different modes of transport. The literature ultimately highlighted the need for context specific information; actual load factors, actual journey times considering traffic/engineering works and so on.

Thus, rather than being explicitly an exercise in answering a specific question, a specific question was used to drive the development of a tool which may hold promise for answering a range of transportation related questions. The original contributions of this work are, firstly the use of real-time data sources to quantify temporally and spatially dynamic network performance metrics (eg. journey times on different transport models) and secondly to organise these data sources in a framework which can handle the volume and type of the data and organise the data in a way so that it is useful for the dynamic agent based modelling of future scenarios.