Optimizing travel routes using temporal networks constructed from GPS data

TL;DR

This paper introduces methods for analyzing urban mobility using GPS data by constructing temporal networks and solving time-dependent routing problems, enabling real-time traffic assessment and efficient route planning.

Contribution

It proposes novel evaluation techniques for mobility using temporal networks and formulates a time-dependent traveling salesman problem for route optimization.

Findings

Temporal networks reveal traffic pattern changes over time.

Time-dependent TSP improves route efficiency considering traffic variations.

Case study with Kyoto data demonstrates practical applicability.

Abstract

Because of the complexity of urban transportation networks and the temporal changes in traffic conditions, it is difficult to assess real-time traffic situations. However, the development of information terminals has made it easier to obtain personal mobility information. In this study, we propose methods for evaluating the mobility of people in a city using global positioning system data. There are two main methods for evaluating movement. One is to create a temporal network from real data and check the change in travel time according to time zones or seasons. Temporal networks are difficult to evaluate because of their time complexity, and in this study, we proposed an evaluation method using the probability density function of travel time. The other method is to define a time-dependent traveling salesman problem and find an efficient traveling route by finding the shortest path. By creating a time-dependent traveling salesman problem in an existing city and solving it, a traveler can choose an efficient route by considering traffic conditions at different times of the day. We used 2 months of data from Kyoto City to conduct a traffic evaluation as a case study.