Machine learning and choice modelling techniques have been developed in parallel as tools for modelling human decisions. While machine learning techniques have gained immense popularity in recent years for their efficiency in utilizing large datasets, the choice modelling techniques are based on theories of economics and psychology and have stronger behavioural underpinning. Hence, the choice modelling techniques may have an advantage in better predicting decisions in radically different future scenarios. The proposed project will contrast these two parallel streams of research and combine them to get the best of both worlds. Emerging big data sources in the context of activity and mobility (e.g. mobile phone and GPS traces, smart cards, etc.) will be utilised in this regard to develop improved models of transport and energy usage. The models will be utilized to evaluate alternative future scenarios.
This 3.5 years EPSRC DTP award will provide tuition fees (£4,500 for 2019/20), tax-free stipend at the UK research council rate (£15,009 for 2019/20), and a research training and support grant of around £5,000.