The Future of Urban Modelling: From BLV to AI

Abstract Urban modelling has a very long history and is widely applied and tested especially in transport and retail analysis, and to an extent with comprehensive models. The theoretical base is fragmented and there are opportunities, sketched here, for more effective integration. There are also opportunities for a much wider range of application. However, with new developments in data science and AI, and in more established but disconnected related areas such operational research, there are opportunities for new theory which will further strengthen the field. ‘Big data’ provides opportunities for real-time model calibration and applications, bigger and better information systems, and the basis for deployment in AI. ‘Machine learning’ can be inverted conceptually to ‘learning machines’ and we show how applications in health, providing augmented intelligence for clinicians, can be translated into urban analytics and planning, offering a potentially important research challenge and new theory.