ANALYTICALLY DERIVED CLASSIFICATIONS OF DAILY TRAVEL-ACTIVITY BEHAVIOR: DESCRIPTION, EVALUATION, AND INTERPRETATION

Abstract

One phase of a research study designed to enhance our understanding of urban travel behavior is discussed. In particular, this paper comprises the description, evaluation, and interpretation of analytically derived classifications of daily travel-activity behavior. A methodology that facilitates systematic identification of groups of similar travel-activity patterns is used to derive the typologies reported and examined here. This methodology is applied to a sample of 236 daily travel-activity patterns drawn from the Baltimore travel demand data set. Each of the sample observations is described by the set of stops made in a 24-h period; each stop is characterized by activity and time of day. The paper explores the hypothesis that a set of daily travel-activity patterns can usefully be described by its membership in a limited number of interpretable general classes. The results reported here show that the 236 daily patterns can be grouped into between 12 and 5 relatively homogeneous groups while retaining between 64 and 46 percent of the information in the data. The results also show that the clusters may be interpreted by defining a small number of representative daily travel-activity patterns for each group. The analytically derived general classes of travel-activity behavior that are described and examined define a categorical response variable for use in models and analyses of daily urban travel behavior.