Ex-Post Evaluations of Demand Forecast Accuracy: A Literature Review

Abstract

Travel demand forecasts play a crucial role in the preparation of decision support to policy-makers in the field of transport planning. The results feed directly into impact appraisals such as cost–benefit analyses and environmental impact assessments, which are mandatory for large public works projects in many countries. Over the last few decades, there has been increasing attention given to the lack of demand forecast accuracy. However, since data availability for comprehensive ex-post appraisals is problematic, such studies are still relatively rare. This study presents a review of the largest ex-post studies of demand forecast accuracy for transport infrastructure projects. The focus is threefold: to provide an overview of observed levels of demand forecast inaccuracy, to highlight key contextual and methodological differences between studies and to highlight key focus areas for future research in this field. The results show that inaccuracy remains problematic for road, rail and toll projects alike, but also how the lack of methodological clarity and consistency calls for a careful interpretation of these results. Mandatory, systematic ex-post evaluation programmes are suggested as a necessary tool to improve decision support, as data availability for ex-post studies is often remarkably poor even for internal audits.

Keywords:

• demand forecasts
• project appraisal
• ex-post evaluation
• inaccuracy
• bias
• decision support

Notes

1. This task has been conducted during an on-going research project, hence the extensive timeframe.

2. The reason for the smaller threshold for rail projects is that most countries construct vastly more road than rail projects, and the available population for sampling is thus much smaller. Large-scale rail studies simply do not exist.

3. I, inaccuracy; O, observation; F, forecast.