This is a study to determine boundaries of station influential area in a residential district which is comparatively built-up. Under this situation, any small district which is chosen fruly from this area is thought to be included in one or more station influential areas. Thus this study means to decide the locations of its boundaries. A method to determine its boundaries from deterministic view point, is to locate any set of boundaries in that area, and to choose specific one showing the most value of fitness. As for the boundary between stations, a datum point will be put on a station itself, and in the case of a boundary between railways, the distance from each line will be chosen as that point. But, as a result of this study, it was proved that to determine the boundary deterministically was difficult. Then it is thought to use a probabilistic model for the boundary determination. Then, it was induced from Fermi-Dirac distribution of statistical mechanics which is not deferministic but a prioristic. It is a generalized model which is applicable on such problems as to follow Fermi statistics and handles with occupancy ratio, related to the alternative judgement between two, and that in the case of permitting to collect many similar energy levels to one group. In that model only two parameters are allowed, which show such specific character as commuters and railways. These parameter are cmportant because of giving explanation for transportation facilities and commuters.
View full abstract