Forecasting, both short and long term, is not a matter of just using statistical techniques; it is important to apply sense checks to forecasts, such as whether the organisation has the resources (financial, human, capital) to achieve the forecasts and also to relate the forecasts to the “real world”. Using railway operations as an example, if a historic high for on time arrivals is forecast, is such as target achievable given external factors which will influence the forecast?
A critical part of the Wigan Transportation Hub Development Study (2001) was to examine the behaviour of passengers using Manchester airport. In particular, what factors influenced the modal choice of passengers (employees were excluded from the analysis as this group as a whole lives significantly closer to the airport than Wigan). The key method used was multilinear regression, examining the modal share of rail from around the top fifty local boroughs into the airport.
The results showed that, beyond a certain car journey time – approximating for a distance for which rail would be competitive over private car and taxi – rail took a significantly larger share only if through services existed.
Other major forecasting projects have included work undertaken for Euromonitor on forecasting travel, transportation and tourism accommodation demand for the United Kingdom.
