Prioritisation of a National Road Infrastructure Programme Using Multi-criteria Analysis

Prioritisation of a National Road Infrastructure Programme Using Multi-criteria Analysis


A Guehnemann, J Laird, A Pearman, University of Leeds, UK


A novel fusion of multi-criteria and cost-benefit analyses is applied to upgrading Ireland's National Secondary Road network. The stability of results is tested, yielding improved transparency and reliability for the decision making process.


Prioritising competing transport infrastructure investment schemes is a common problem at national, regional and more local levels and often the number of potential schemes that has to be considered is substantial. With pressure on resources, growing awareness of the range of scheme impacts that has to be taken into account and public pressure to be able to demonstrate a transparent, consistent and effective ranking process, decision making bodies require comprehensive and reliable appraisal methods. In most European countries, a cost-benefit analysis (CBA) or multi-criteria method (MCA) is applied in the appraisal process. MCA has a number of advantages over CBA. Primarily it allows impacts that cannot be monetised or cannot be quantified easily to be included in an appraisal in a transparent way. Such impacts could include those on sites of outstanding beauty or of historic importance. A general criticism of MCA is that the rankings are dependent on the weights given by decision-makers' preferences and not generally accepted measures of social preference. Thus, weights often appear arbitrary and can be subject to change. Another shortcoming of MCA is that generally decision-makers require measures for value of money as well. Hence, in many countries the two methods are implemented alongside each other, but often with no synthesis of results into one joint framework. This leaves the final decision about the worthiness of investments to the decision-makers, but is open to interpretation and criticism.

This paper presents an MCA methodology that addresses these challenges in an innovative way. It was developed for Ireland's National Roads Authority (NRA) and has been applied to a study examining the upgrade of Ireland's National Secondary Road network. This consists of 2,700 km of primarily rural single carriageway 'A-roads' that complement the national primary routes. From a policy and political perspective it is necessary to direct investment to those parts of the network where it is needed the most and to ensure the programme justifies the investment costs. MCA is a key instrument in delivering this. A specific innovation here was to develop a methodology that incorporates CBA elements in the MCA framework in a consistent manner and to test the stability of results. A further innovation is the method used to test the sensitivity of the rankings to variations in weights, as such weights can be quite reasonably contested by different stakeholder groups. A key message of the paper is that the methods can be introduced to make a national investment strategy developed through an MCA robust to the usual criticisms levelled at MCAs without losing its key advantages.

In order to derive an overall appraisal measure that allows a comparison and prioritisation of projects, a scoring and weighting framework has been developed using a linear additive approach. In total, 566 projects have been scored and ranked accordingly. A BCR threshold was then applied to calculate the MCA score that projects needed to achieve to provide value for money. The so defined investment programme of worthwhile projects has been further condensed by applying an incremental analysis to decide between mutually exclusive projects.

From the point of view of the decision maker it is important to have an understanding of the robustness of the size of, and rankings within, investment programmes created. For this reason, a variety of sensitivity tests have been devised and applied to explore the extent to which the initial rankings may be affected by changes in view about the relative importance of the different impacts.

Based on this analysis, there is no strong evidence of very high sensitivities in the rankings that the assessment has constructed against changes in weights. While the size of the programme is sensitive to changes in weights, schemes that come towards the top of the ranking are good performers in terms of the criteria and the relative importance of those criteria that have been assessed in collaboration with the decision makers, and reasonably sized variations in the data do not induce drastic changes in the rankings obtained. Even under more radical changes in the structure of weight sets, the best performing schemes stay at the top of the rankings. This result gives a degree of assurance for the decision makers to proceed with investment in the best schemes in the shorter term and to draw up an indicative programme for future investment for the period after that.


Association for European Transport