Traffic crashes are a major cause of death in both developed and developing countries. Traditionally, most of the responsibility for such records was attributed to drivers, but the repetition of the cases in specific regions would support the hypothesis of the existence of higher risk locations. In a general understanding, this identification would have a fundamental importance in attempting to mitigate this serious problem. Since this would allow greater efficiency in the application of safety measures to reduce the probabilities, or even the severity, of the occurrences. The aim of this thesis was to investigate the use of the jerk (the first derivative of acceleration in time) as an indicator of runoff road risk locations caused by lack of geometric consistency between the approach tangents and the subsequent horizontal curves. For this purpose, a controlled experiment was carried out in a fixed-base driving simulator. The results of the research showed that there was a significant effect of the variation of the geometric characteristics in the value of the applied average minimum jerk. There was still a strong correlation between the proposed indicator with two traditional methods. The results would encourage the further development of the use of the minimum jerk as an indicator of geometric consistency for highways. In times of Big Data and sensors spread out throughout the population, the use of a seemingly simplified obtainable variable would allow the real-time survey of infrastructure problems, such as the geometric inconsistency of the roads, with the contribution of the users themselves.