The knowledge of the actual operating speed of free vehicles all along a highway (the speed profile) is of prime interest in safety studies. However, it is not easily obtained, for it needs, with traditional methods, a considerable number of jopurneys with a dedicated equipped vehicle. Starting from the knowledge of the actual 85th operating speed profiles on some rural highways, obtained in such a way, is evaluated the accuracy of an alternative method, combining in an elementary way few individual speed profiles and few spot speed measurements. An empirical estimation of the resulting error is given, for different numbers of continuous and spot speeds measurements. It appears that with four individual continuous speed profiles and four spot speed distributions, a quite reasonable accuracy may be expected. If such a method is less pertinent than spot speed measurements for studying given points of a road, and not very useful ing to know the speed profile along only a single highway, it may be useful to get an idea of the actual speed behaviour on each point all over a secondary rural network. (A). For the covering abstract of the conference see E217780.
Abstract