jats:p Masonry arch bridges, most of which have far exceeded modern design lives, have demonstrated themselves to be sustainable structures with low life-cycle costs. However, increased traffic loading and material deterioration over time necessitate periodic reassessment of these structures. There are numerous different analytical methods available for the assessment of masonry arch bridges. The expectation is that for increasing levels of assessment complexity an increase in load capacity converging on the ultimate capacity would be achieved. In this paper it is demonstrated that this is not always the case. This has cost implications for both the bridge assessment itself and for costs associated with load restrictions and strengthening measures. Five different assessment methods were selected to assess a set of 11 single-span bridges, ranging in span from 2·4 m to 15·2 m, with the objective of reviewing and rationalising current assessment guidelines for masonry arch bridges. The bridges chosen are a representative sample of the stone arch bridges on the Irish National Roads network. It was found that there is a significant variation in assessed capacity depending on the assessment method used. Limit state analysis methods were found to generally result in higher ratings for segmental bridges while elastic methods resulted in higher ratings for three-centred or semi-circular bridges. Assessment ratings found using the Military Engineering Experimental Establishment (MEXE) method were difficult to rationalise across the bridge set considered in this study. Following a review of the origins of the MEXE method and its current form as set out in the assessment guidance, it is recommended that its use as the predominant tool in a simplified assessment procedure is not appropriate and that a more rational approach is required for a more realistic and reliable calculation of bridge capacity. The development of an improved assessment methodology is being considered as part of the current study. </jats:p>