柔軟な距離減衰関数を用いた都市施設のアクセシビリティ分析 ANALYSIS ON ACCESSIBILITY TO URBAN FACILITIES USING THE FLEXIBLE DISTANCE ATTENUATION FUNCTION

In this study, we aim to clarify spatial distributions of facility accessibility that vary according to facility type by using a distance attenuation function that includes an unknown parameter in the determination of distribution function. The distribution of facility accessibility is represented as a monotonically decreasing distance function that includes two parameters indicating “distance resistance” and “type of distribution function.” Then, the sum of the square of the difference between population density and facility accessibility is minimized. We have applied the distance function in a defined region and analyzed facility accessibilities from multiple perspectives.

In this study, we aim to clarify spatial distributions of facility accessibility that vary according to facility type by using a distance attenuation function that includes an unknown parameter in the determination of distribution function.The distribution of facility accessibility is represented as a monotonically decreasing distance function that includes two parameters indicating "distance resistance" and "type of distribution function."Then, the sum of the square of the difference between population density and facility accessibility is minimized.We have applied the distance function in a defined region and analyzed facility accessibilities from multiple perspectives.(1)

Keywords :
(1) (1) d (2) (2) (4) (4) From the perspective of maintaining a balance between service supply and demand, the spatial relationship between a residence and a facility in urban areas is an important issue.In urban planning, variations in accessibility to facilities that change with distance are often regarded as monotonically decreasing distance functions.These functions signify that spatial availability reaches its peak at each location of facilities and decreases as distance increases.In such cases, the value of power in distance function is generally determined in advance.However, it is difficult to select an appropriate function type that can be used as the base for facility accessibility.In this study, we aim to clarify spatial distributions of facility accessibility that vary according to facility type by using a distance attenuation function that includes an unknown parameter in the determination of distribution function.
One way to estimate distance attenuation function for the analysis of facility accessibility has been discussed in this paper.As for the distribution of population corresponding to facility accessibility, population densities are calculated by dividing population by area using aggregated data.The distribution of facility accessibility is represented as a monotonically decreasing distance function that includes two parameters indicating "distance resistance" and "type of distribution function."Then, when the sum of the square of the difference between the two distributions over the entire study area is minimized, the minimum values are derived as estimates of each parameter.Next, spatial distribution of facility accessibility by type of facility is clarified using the estimates of the two parameters, based on the distance functions.
In this study, we have applied the distance function in a defined region and analyzed facility accessibilities from multiple perspectives.The study area consists of 12 elementary school districts situated in the central area of Oita City, Oita Prefecture, in Japan.We selected 12 types of facilities, such as convenience stores, medical institutions, post offices, elementary schools, and general merchandise stores, for analyzing facility accessibilities.From the results, many estimate values of the parameters that represent the values of power in the distance functions were close to 1 or 2. It suggests that the use of exponential distributions and normal distributions in previous studies on facility accessibilities is reasonable, as far as the result of this study is concerned.The types of facility are classified into four groups by cluster analysis, based on the estimates of the two parameters of the distance function.In addition, we calculated the relative error between density of population and facility accessibility, and visualized their spatial distributions on the map.
The values of power in the distance functions, which are estimated in this study, are expected to be utilized for not only evaluating the estimates themselves but also as guidelines for applying representative distributions, such as normal distributions, to distance functions.To formulate a facility accessibility in this study, only the nearest neighbor facility from the residence has been considered.However, facilities other than the nearest neighbor facility must be considered if there are opportunities to choose other facilities or when there are capacity constraints of facilities.To address this problem, one of the solutions is to formulate accessibilities with a parameter that has a certain degree of priority in use in more than one facility.

ANALYSIS ON ACCESSIBILITY TO URBAN FACILITIES USING THE FLEXIBLE DISTANCE ATTENUATION FUNCTION
Yusuke KATAOKA ＊ ＊ Assoc.Prof., Dept. of Management, Faculty of Contemporary Business, Kyoto Tachibana University, Ph.D.
(2017 年 4 月 1 日原稿受理，2017 年 9 月 25 日採用決定) ─ 82 ─ ., Dept. of Management, Faculty of Contemporary Business, Kyoto Tachibana University, Ph.D. ANALYSIS ON ACCESSIBILITY TO URBAN FACILITIES USING THE FLEXIBLE DISTANCE ATTENUATION FUNCTION * Yusuke KATAOKA Masuyama, A.:The Relationship between the Geographic Distribution of the Elderly in Urayasu-shi and the Accessibility of Medical and Commercial Facilities for the Elderly, Journal of the City Planning Institute of Japan,Vol.42,No.2,pp.72-79,2007.10(in Japanese): Masuyama, A.: A Restricted Randomization Test for Analyzing the Relationships between the Population Distribution in a Region and the Accessibility to Daily Facilities, Journal of the City Planning Institute ofJapan, Vol.45, No.3, pp.583-588, 2010.10 (in Japanese) Assoc.Prof., Dept. of Management, Faculty of Contemporary Business, Kyoto Tachibana University, Ph.D.