p. 669−686
2345-6922
Vol.42/No.4
p. 687−707
2345-6922
Vol.42/No.4
p. 709−718
2345-6922
Vol.42/No.4
p. 719−733
2345-6922
Vol.42/No.4
0/5 and Bartlett's Test <0/05) in the F'ANP, 5 indicators include the share of commercial land use, religious, green space and gardens, river existence and affordable housing from the research process has been deleted. In the first stage of implementation of the two-stage F'ANP model for the implementation of the Factor Analysis model number of factors, percentage of change and the load factor of 9 indicators including the share of educational land use, health, law- office, cultural- arts, strength of building, facilities and infrastructure, sports, sewerage coverage and access to public transport have been calculated. In the second stage of F'ANP model that includes ANP model, after the formation of the super matrix (Network model) and calculating the Matrix elements in Excel and MATLAB software, The relative importance of the indicators has been derived and finally To achieve the output of research these coefficients have been multiplied In 15 areas and its results have been shown in the map in GIS software. Discussion of Results The model used in this study is F'ANP that it has been presented for the first time in 2013 by Esfandiar Zebardast for manufacturing composite indicators. The F'ANP model process includes two-stage, which its first stage is Factor Analysis. Accordingly, in order to assess the livability in 15 areas of Urmia, first according to existence of information, number of14 indicators include the share of commercial land use, religious, green space and gardens, river existence, affordable housing, educational land use, health, law- office, cultural- arts, strength of building, facilities and infrastructure, sports, sewerage coverage and access to public transport have been selected and their amounts have been calculated in SPSS software. In the first stage of implementation of F’ANP model in order to calculate the Factor Analysis of indicators in SPSS software, 5 indicators were removed from 14 primary indicators so that appropriateness of Factor Analysis of the indicators comes (KMO=0/52 وTest=0/04 Bartlett’s). To determine the number of factors that have to be extracted in the analysis for the sets of datas, Kaiser Criterion was used. According to this criterion, the only factors with eigenvalues of 1 or more are accepted as possible reference of changes in data and the Factor has the highest priority that has the most eigenvalues. When the Factor Analysis was done by Varimax rotation, clear structure of factors with 4 indicators were done that totally explain about 85.2% of the total variation. In the second stage, the network analyze process is being used. According to obtained results of Factor Analysis, network model for determination the livability has been formed. In this diagram first cluster is the goal of research, the second cluster is the livability dimensions and the third cluster include the subset of each dimensions is derived from Factor Analysis. Forming the indicators of each dimensions are related to each other. After building the network model, the super matrix has been composed and the individual matrix of it will be built. In this matrix [W_21] vector show the relative of the research goal and 4 dimensions of livability. So, to calculate the [W_21] vector like the conventional process in ANP, the binary comparison between the four dimensions in order to achieve the goal of research should be done. Namely in the construction of binary comparison of [A_21] matrix, instead of using the Sati’s Quantitative 9 scale, the percentage of changes that each factor explains, is used. For calculating the importance of 4 factors, first the geometric mean of the [A_21] matrix of row elements earns and then normalize them so that [W_21] vector is obtained. The elements of [W_32] matrix, show the relationship between factors and indicators. In F’ANP model variable loads, are considered as well as their importance in binary comparison of [W_21] matrix binary. Since it was shown, instead of the formation of a binary comparison matrix importance coefficient can be obtained directly from the relevant vector normalization, so the Weight of elements vector related to the first factor (X_(1 )) is achieved through normalizing the load factor of indicators. The elements of [W_33] Matrix, show the interdependence between the constituent indicators of each factors. To obtain the importance coefficient of constituent indicators of the prime factor (X_(1 )) the correlation coefficients matrix between these indicators has been obtained and are normalize them. The importance coefficient of constituent indicators of other factors is calculated similarly. Thus, the [W_33] matrix, is calculated. After calculating the elements of basic matrix, we replace them in the basic super matrix to obtain issue super matrix.obtained super matrix is weighted (sum of columns is equal to 1), then bring it to the limit until obtain the relative importance coefficient of indicators. Important coefficient of indicators is recoverable from goal column in limited super matrix ; the vector is normalized to obtain the relative importance of the indicators. In the next stage any of indicators explain a subject matter, after determination and action the relative importance weight for each of them are combined together that in this article the Arithmetic procedure is used. After the relative importance coefficient of livability indicators were obtained, by calculating the SOVI_i composite indicators of livability have been achieved in the 15 areas of Urmia city. Conclusions The results of applying the F’ANP model to assess the livability in the 15 areas of Urmia show that 158 hectares of land areas of Urmia city contains 9% in the too low livability zone, 696 hectares contains 38% in the low livability zone, 457 hectares contains 25% in the average livability zone, 250 hectares contains 13% in the high livability zone and 281 hectares contains 15% in the too high livability zone has been located.]]>
p. 735−752
2345-6922
Vol.42/No.4
p. 753−769
2345-6922
Vol.42/No.4
p. 771−785
2345-6922
Vol.42/No.4
p. 787−803
2345-6922
Vol.42/No.4
p. 805−822
2345-6922
Vol.42/No.4
p. 823−838
2345-6922
Vol.42/No.4
p. 839−854
2345-6922
Vol.42/No.4
p. 855−867
2345-6922
Vol.42/No.4
p. 1−35
2345-6922
Vol.42/No.4