Efficient Geo-Computational Algorithms for Constructing Space-Time Prisms in Road Networks

Abstract

The Space-time prism (STP) is a key concept in time geography for analyzing human activity-travel behavior under various Space-time constraints. Most existing time-geographic studies use a straightforward algorithm to construct STPs in road networks by using two one-to-all shortest path searches. However, this straightforward algorithm can introduce considerable computational overhead, given the fact that accessible links in a STP are generally a small portion of the whole network. To address this issue, an efficient geo-computational algorithm, called NTP-A*, is proposed. The proposed NTP-A* algorithm employs the A* and branch-and-bound techniques to discard inaccessible links during two shortest path searches, and thereby improves the STP construction performance. Comprehensive computational experiments are carried out to demonstrate the computational advantage of the proposed algorithm. Several implementation techniques, including the label-correcting technique and the hybrid link-node labeling technique, are discussed and analyzed. Experimental results show that the proposed NTP-A* algorithm can significantly improve STP construction performance in large-scale road networks by a factor of 100, compared with existing algorithms.

Understanding the Impacts of Human Mobility on Accessibility Using Massive Mobile Phone Tracking Data

“…They are expressed, respectively, as cumulative lengths of accessible links (Kwan 1998) and cumulative activity durations on all accessible links (Neutens et al 2008;H.-P. Chen et al 2016). In addition to these 2D and 3D indicators, traveled distance along the user's trajectory during three time periods of interest is also calculated, as it is a commonly used one-dimensional (1D) indicator in the literature.…”

Measuring the structural similarity of network time prisms using temporal signatures with graph indices

“…Existing NTP models represent a road network as a graph where its edges are straight lines connecting vertices in the two‐dimensional plane (Chen, Chen, Wang, & Li, ; Kuijpers & Othman, ). The NTP models then apply single‐source shortest‐path algorithms (e.g., Dijkstra's algorithm and A* algorithm) to determine the graph of PPN: .…”

“…Algorithm 1 presents the computation procedure of the temporal sweeping. This algorithm requires the graph of PPN as input, and we employ the NTP‐A* algorithm which has proven more efficient in a comparative experiment for constructing PPNs on a large‐scale road network in this article (see Chen, Chen, et al, for the detailed procedure of the algorithm to construct PPNs).…”

Exact space–time prism of an activity program: bidirectional searches in multi-state supernetwork

“…To further narrow down the search space, Chen et al (2016) proposed a one-way goaldirected TBS geo-computational algorithm. Specifically, A* search (Doran 1967) is first used in the forward shortest path search from H 0 .…”