In this paper, an experimental study was carried out in various road sections by means of an original methodology: video-camera recordings were acquired and data was processed by means of a semi-automatic procedure in order to obtain the speed and the position of every vehicle. The research was preceded by a first phase where measurements have been taken and evaluated for a better knowledge of the conditioning effects on the users. Afterwards, various devices and special arrangements of signalling have been experimented in a given road section. By doing so, it has been calculated, at each installation, the modifications induced on the vehicles motion, in terms of speed and transversal position, under various environmental and temporal conditions (day and night lighting). A straight road section was studied, during three different traffic flows, under four lighting and four weather conditions. We have worked on a total of six thousand vehicles, comprising four different typologies. Results show the effectiveness of different tested configurations, and indicate, under each environmental and temporal condition, the most appropriate equipment for a better management of the vehicular speed and transversal position. Analysing the data collected during this research, in which were experimented some measures to modify positively the driver's behaviour, it was evident that this is possible but, sometimes, the observations are partially different from projections. The driver's behaviour, as for the transversal dislocation and the speed, was modified consequently to the present or installed elements. In particular, the installation of lateral tape produced some effect on the dislocation and velocity, but that is probably due to the "surprise" effect of such an eye-catching and anomalous element. It was not possible, due to the low durability of the tape, to examine the long term effect. The optical strip gave the best results during the night drive, expressly in the dislocation. These elements, in fact, are an aid for a correct evaluation of the dimensions and the position of the lane. The variation in speed produced by this operation are not clear. The rumble strips modified the dislocation and, mostly, reduced speed, in the daytime too, that was a condition in which the other operations failed. The intervention was really effective, and is quite good in the long term. Given that these strips were installed in a straight section, in which there is no reason to slow down, even a little speed reduction (by means of 4-7 per cent) could be a good result: the effect should be better in presence of dangerous curve, crossing, etc. It has to be reported that, after some months of installation, some small cars began, during the night, speed competitions in the examined section. The installation of a enlarged median was the most efficient operation, either for dislocations or for speed. The effectiveness was stable during the time and there was a good reduction of illegal overtaking. That is an important indirect benefit, even with long straight roads. (A) For the covering abstract of the conference see ITRD no 207828. The reprints are also available at the web - http://www.vti.se/pdf/reports/K18APart1.pdf; http://www.vti.se/pdf/reports/K18APart2.pdf and http://www.vti.se/pdf/reports/K18APart3.pdf.
Abstract