An approach for evaluating the effectiveness of traffic guide signs at intersections

•In this study, a generic approach is developed to examine and evaluate the effectiveness of traffic guide signs using simulation experiments.•A traffic guide sign evaluation method (TGSEM) is developed and illustrated using examples of traffic guide sign schemes in suburban Beijing.•Based on subjective experiments, the ergonomic evaluation model (DCI, the abbreviation of demand, comprehension, and information volume) was obtained.•During driving simulation experiments, seven indicators were analyzed: average speed, standard deviation (SD) of speed, average acceleration, standard deviation of acceleration, travel time, braking frequency and throttle power.•Then, through the grey relational analysis of subjective and objective indicators, the optimal scheme of four traffic guide signs is obtained. Traffic guide signs play important roles in people’s daily lives. However, the effectiveness and performance of traffic guide signs at intersections are significantly impacted by many factors, such as the types of information on traffic signs, their information volume and comprehensibility, the behavioral attributes of drivers, the geometric features of roadways, and weather and visibility conditions. When deploying traffic guide signs, efforts are needed to clarify whether the installation of a traffic guide sign is warranted. In this study, a generic approach is developed to examine and evaluate the effectiveness of traffic guide signs using simulation experiments. A traffic guide sign evaluation method (TGSEM) is developed and illustrated using examples of traffic guide sign schemes in suburban Beijing. The questionnaires showed that most drivers feel that the current traffic guide signs in suburban Beijing are insufficient and need to be rectified. Then, simulation experiments were conducted. Based on subjective experiments, the ergonomic evaluation model (DCI, the abbreviation of demand, comprehension, and information volume) was obtained. Of the four schemes, scheme 3 was shown to be the most popular. During driving simulation experiments, the analyses of average speed, standard deviation (SD) of speed, average acceleration, standard deviation of acceleration, travel time, braking frequency and throttle power showed that scheme 2 had a better impact on drivers’ behavioral data. Finally, Grey relational analysis showed that scheme 2 has the highest degree of correlation and can be recommended to traffic management departments. The experimental tests and analy