Comparison between an isopleths-based regionalization method nd other methods that yield spatially contiguous sub-regions

Abstract

This paper compares an isopleths-based regionalization method to other methods that yield spatially contiguous sub-regions and investigates whether the isopleths-based method outperforms the other methods particularly in terms of the homogeneity of sub-regions. The Monte Carlo experiment revealed that the isopleths-based method tends to yield more homogeneous sub-regions than the other methods, but in some instances the isopleths-based method underperforms the other methods. However, the experiment also revealed that the more the number of spatial units, the less the cases where the isopleths-based method underperforms.