Even though the crash risk in tunnels is rather lower than on the open road network, crash consequences can be very severe. Therefore it is of high importance to assure high safety standard in tunnels, which includes, but is not limited to, an appropriate illumination. Too low illumination levels may compromise safety, while unnecessarily high levels waste resources. In this study it was investigated in which way different levels of illumination and brightness of the tunnel walls influence the behaviour of attentive and inattentive drivers. The study was conducted in the Simulator III at VTI, which is a high-fidelity simulator with linear motion in lateral direction. A within-subjects design was employed, and 24 participants took part in the test. Illumination was varied on three levels and tunnel wall colour and driver attention were varied on two levels each. This yields twelve different combinations, which meant that each participant drove through twelve tunnels with a length of 4 km. The order was randomised with certain restrictions between participants. The tunnels were separated by 2 km long open motorway sections. In each tunnel the participant encountered an overtaking situation, which demanded a decision on the tactical level. The collected data consisted of driving behaviour data as logged from the simulator, eye tracking data, secondary task performance data, subjective demand ratings while driving, and subjective ratings and comparisons post-test. The data collected during driving were analysed for the situation with the overtaking event and in addition for an an event-free driving situation further down the tunnel. The main results were: • Bright walls are more important for experienced safety and comfort than high illumination levels, as long as the illumination is sufficiently bright. • Driving behaviour and gaze behaviour were heavily influenced by driver state, with distracted drivers showing more unsafe behaviour. • A larger percentage of long glances away from the forward roadway and lower attention levels were found for tunnels with dark walls. • The participants experienced a higher task demand during the distracted condition as compared to non-distracted. They also experienced the overtaking event as more demanding than driving freely. • Bright walls received slightly lower demand ratings. • Secondary task performance did not vary across conditions. It can be concluded that bright walls enhance traffic safety and are conducive to the drivers’ to feel safe and comfortable, provided an acceptable level of illumination. Absolute levels are difficult to come by based on a simulator study, as the range of illumination is much more limited in the simulator than in reality. For detailed results on absolute illumination levels it is indispensable to run real world trials, even though the simulator proved useful for the evaluation of relative levels. As performance was degraded for distracted drivers, it is recommended to address the issue of attention in future studies, because inattention and monotony are known risks especially in longer tunnels. It is also recommended to evaluate design features related to visual guidance, and to investigate other architectural measures, which might help decrease energy consumption without compromising traffic safety. (Author/publisher) This report may be accessed by Internet users at http://www.vti.se/EPiBrowser/Publikationer%20-%20English/R424A.pdf
Samenvatting