Walk this wayAssessing variables of spatiotemporal behavior of pedestrians on a national scale

By Julia Sipkema

Abstract
People travel from one point to the other often multiple times a day, using different modes of transport. Walking as a mode of transport is a large part of these daily movements of people because it is a cheap, nature-friendly, and healthy type of mobility. This is especially relevant regarding current issues like population growth, climate change, and increasing urbanization. Therefore it is important to model pedestrian intensities, showing where and how people walk and how this will be impacted by changes in urban fabric or other types of transportation. However, walking is also a mode of transport that is often forgotten in research and traffic modeling. This is often due to a lack of data available with pedestrian research historically relying on counts and surveys. Furthermore, when pedestrians are included it is often focused on a small extent, like a neighborhood or city, due to this data availability. This research shows that it is possible to estimate pedestrian intensities using the extent of a whole country, which can be used in pedestrian modeling. This was done by analyzing GPS panel data looking at both spatiotemporal and geographical patterns and comparing it with existing literature. Patterns were found in pedestrian intensities that can be seen even on the scale of a province and a country. This includes the fact that people tend to walk more often in dense areas (98 vs 85 trips on average in a month), but they walk shorter distances (1094 meters per trip on average compared to 1180). Furthermore, it was found that people on average undertake 1.07 trips by foot per day with an average of 1.1 kilometers. The research also shows that there is a temporal difference in walking behavior: people walk the most on Saturday, 1.22 trips, compared to an average of 1.04 on other days and people walk longer distances on Sundays (1419 meters per trip on average, compared to a 1088 average on other days). This research shows that although there are differences in the extent and research methods, the results found are comparable to earlier research and national mobility reports. This makes GPS mobile phone data a trustworthy source of data. It also shows that using a large data set can give insight into regional differences. The results show patterns in people's walking behaviors that can be used to create traffic models as well as useful coefficients to finetune them. This research can be a starting point for more precise pedestrian traffic models. Furthermore, it can be seen as a basis for more research on mobility using large datasets, for example using mobile phone GPS data. Finally, this research can be used as an impulse for more research on pedestrians, as they are often underrepresented in mobility research, even though it is such an important type of mobility looking at climate change, crowding, and health.