The driver, through sight, acquires a lot of information from the road environment, most of which is necessary for his safe route. However, if the amount of information per unit of time is excessive, potentially dangerous situations of overload could be created. Even the opposite condition, that of a road that does not adequately stimulate the cognitive functions of the driver, may pose certain safety problems because it triggers the so-called boredom effect. This phenomenon, generally classified under the name of workload, was treated with great depth in literature but, probably, sufficiently detailed methodology has not yet been proposed for making forecasts on this variable along the road. The difficulty of preparing a reliable model can be explained by some of the characteristics of the road environment: many uncertain variables, including the human factor, choosing the most appropriate analytical method, lack of appropriate databases. The purpose of this article, therefore, is to present a prediction model based on the analysis of physiological workload by means of head-eyes movements and fuzzy techniques applied to a real context. The results obtained, although limited by the observed data set, allowed for the prediction with some accuracy of the tendency of the workload, referring also to the overload and under load thresholds position of which was defined on the basis of performance measurements along the road under consideration. In the first stage of the study the methodology is applied to the design of maintenance on an existing road, but once the correctness of the procedure is established, it can also be extended to new roads.