Abstract
The author uses the concept of entropy from information theory to test the application of Christaller's hexagonal trade-area concept to actual settlement patterns. The entropy measure is used to separate the settlement pattern into its hexagonal and random components, just as incoming radio signals would be separated into the original outgoing signals and noise interference. Medvedkov derives formulas for two methods, one using the density of points within cells of a lattice, the other using the nearest-neighbor technique. The entropy approach is tested in seven actual situations, including three control situations, and shows the regular, or hexagonal, pattern to be the predominant component in five out of seven cases.

This publication has 2 references indexed in Scilit: