The question of whether or not traffic load repetitions should be considered in pavement design, and if so by what relationship, is considered. From reports covering widely different experiences it is concluded that a continuum of pavement response to traffic has been described. Local design practice together with traffic and environmental factors determine the sensitivity for any particular case. Comparisons between dimensions for roads of similar specification in Sweden, the UK and several states of the USA suggest that Swedish pavements might be rather insensitive to traffic loads. A method for treating mixed traffic is required for the development of a pavement design traffic index. The equivalent wheel load concept, with equivalency based on equal decrements of serviceability, has been widely used for this purpose in the USA. This approach is justified through mathematical models relating pavement thickness to load applications. Such models reflect specific conditions and pavement design methods and so must be validated before any general adoption of load equivalency factors developed elsewhere can be considered. A method for such validation based on experience in Minnesota is described.