Journal of Korean Society of Transportation ISSN:1229-1366(Print) 2234-4217(Online) 대한교통학회지

As interest in quality of life improvements and sustainable development has increased, societal need for walking expediency promotion is rising. In order to effectively promote walking related projects and policies, an understanding of pedestrian behavior must take precedence. However, information including, among others, the types of walking environments in which pedestrians feel uncomfortable and thus alter their existing walking speed and direction, and the types of pedestrian attribute groups that are particularly sensitive to walking environments is still insufficient. In this study, we analyze conditions under which the pedestrian changes their existing walking speed and direction. We also examine whether the tendency of this selection is different between pedestrian attribute groups. In order to achieve this, the research is conducted using a stated preference survey based on discrete choice experiments, asking for a preference between a maintenance or change in walking velocity in a virtual walking environment. Influences on changes in walking velocity is determined using the binary logit model for the following factors: (1) the social walking environment, including attributes such as distance, angle, relative speed, walking density with respect to another pedestrian or obstacle, (2) the physical environment regarding the width of the walking space and its slope, and (3) individual pedestrian attributes such as age, gender, trip purpose, and trip frequency. The result show that closer distance to the obstacle or other pedestrian, obstacles positioned in front than to the side, slower relative speeds of the obstacle or other pedestrians and narrower width of the walking space have statistical significance in affecting the change in walking speed and direction. Other findings from this research, we found that older people have lower likelihood of changing their walking speed and direction. Compared to non-business purpose trip pedestrians, the likelihood of pedestrians on business-purpose trips changing their walking velocity was higher. There was also a clear difference between genders. On the other hand, there was no statistical significance between differences in selection tendency of walking velocity according to congestion level, slope, and pass frequency in the walking space. With these findings, we first provide insights on how to evaluate walking policy effects and service levels for individual pedestrians from a micro-level perspective. For example, the possibility that a pedestrian changes velocity or the expected number of changes in velocity can be taken into consideration as an evaluation index. Secondly, the fact that results of this research can be utilized as basic data for the development of a walking behavior simulator at the micro-level will increase its importance and utilization factor in the future.