Modelling commuter patterns: a spatial microsimulation approach for combining regional and micro level data

Transport to work is a universal phenomenon, but is uneven over space. The distribution of mode and distance statistics vary depending on a range of factors. The scale of analysis (from individual to local and regional levels), the location of the study area (e.g. urban or rural settlements), and the socio-economic characteristics of the target population all influence commuter patterns. This heterogeneity is problematic for decision makers tasked with encouraging more sustainable and less costly commuter patterns based on transport to work statistics. Existing studies on commuting fail to consider the multiple levels at which transport systems operate, and leave important questions unanswered. For example, should policies target individuals, local areas, or regions, and at what level should they operate for maximum benefit? This paper outlines this research problem, and then describes an approach for tackling it based on a case study of Yorkshire and the Humber, an economically peripheral region of the UK. Our spatial microsimulation model uses iterative proportional fitting (IPF) to simulate the characteristics of individual commuters, in terms of socio-economic class, age, sex, and income, while geographically aggregated commuter statistics are constrained by Census data. This represents a novel application of spatial microsimulation, to model commuter behaviour. The approach has the potential to allow 'what if' scenarios to be undertaken, and opens up the possibility of dynamic microsimulation to policy makers. This latter possibility is attractive because it allows scenario-based projections of the future for the evaluation of policy assumptions. The results of the static model illustrate the importance of accounting for variability at the individual level when devising transport policies. The method is discussed with respect to long term transport policy objectives at the EU level. In conclusion, our approach could provide valuable information for policy evaluation at individual, local and regional levels. Keywords: spatial microsimulation, commuting, transport policy JEL Code: R49