Archived travel data (ATD) derived from various private sources has attractive characteristics that suggest it can be a suitable replacement for traditional sample-based external travel studies or other similar origin-destination (OD) studies. The hope is that this new source of information will reduce or eliminate several negative characteristics of traditional methods. Before this hope can be realized, however, the new solution must be validated; this is the main intent of this research project.

The space-time prism (STP) is a central concept in time geography, demarcating the geographic limit of an object's potential movement with respect to time. The STP is widely used to measure individual accessibility to opportunities in space, estimate potential exposure to the environment and capture space-time path uncertainty based on sampled locations of a moving object. Similarity measures for movement data play a fundamental role in the knowledge discovery process from large moving object databases; however, there has been little attention to such similarity measures for STPs and its variants. The goal of this dissertation is to develop similarity measures for two types of more realistic representation of the planar STP: network-time prisms (NTPs) and field-time prisms (FTPs). The main idea of the prism similarity measure is to reduce the dimensionality of the three-dimensional prisms by recording a selected prism property with respect to time, generating a one-dimensional array known as the prism’s temporal signature. This approach is advantageous in that existing path similarity measures for movement trajectories can be applied to assess the similarity between the temporal signatures. The method can also measure similarity not only in the geometric or structural prism properties but also with respect to semantic properties. The first study develops and demonstrates the temporal signature approach for similarity analysis of the structural properties of NTPs. Graph‐theoretic indices play an essential role in summarizing the structural properties of NTPs at each moment. Two extensive simulation experiments demonstrate the feasibility of the approach and compare graph indices' performance for measuring NTP similarity.