Multidimensional Sequence Alignment Methods for Activity-Travel Pattern Analysis: A Comparison of Dynamic Programming and Genetic Algorithms

摘要

Quantitative comparisons of space-time activity-travel patterns have been made at length in regional science. Traditionally, Euclidean Hamming distances have been widely used to measure the similarity between activity-travel patterns that involve several attribute dimensions such as activity type, location, travel mode, accompanying person, etc. Other techniques, such as pattern recognition in signal processing theory, have also been introduced for this purpose. All these measures, however, lack the ability to capture the sequential information embedded in activity-travel patterns. Recently, the sequence alignment methods (SAMs), developed in molecular biology that are concerned with the distances between DNA strings, have been introduced in time use research. These SAMs do capture the similarity of activity-travel patterns, including sequential information, but based on a single attribute only. Unfortunately, the extension of the unidimensional SAMs to a multidimensional method induces the problem of combinatorial explosion. To solve this problem, this paper introduces effective heuristic methods for the comparison of multidimensional activity-travel patterns. First, following a brief review of existing measures of activity-travel pattern comparison, the problem of multidimensional sequential information comparison and the combinatorial nature of the method are discussed. The paper then develops alternative multidimensional SAMs employing heuristics based on dynamic programming and genetic algorithms, respectively. These heuristic SAMs are compared using empirical activity-travel pattern data. The paper ends by discussing avenues of future research.