Clustering Permukiman Kumuh Di Kabupaten Siak (Studi Kasus: Kecamatan Tualang, Kecamatan Minas, Kecamatan Kandis)

The high rate of population growth in Siak Regency causes the need for housing to increase. However, these needs are not matched by the availability of land and affordable housing, especially for middle to lower income people. This condition triggered the emergence of slums, scattered in Siak Regency. In general, the impact caused by the existence of slum areas has an effect on the condition of occupants, housing, and infrastructure. This study aims to classify slums based on their similar characteristics. The sampling technique was carried out using the proposive propotional random sampling technique. The type of sample used is proposive sampling technique with certain considerations, with special selection and applying the characteristics in accordance with the objectives. There are 2 analytical tools used in this study. First, simple qualitative descriptive to identify the characteristics of slums based on conditions of occupants, housing and facilities. Qualitative descriptive analysis is done by processing questionnaire results into Likert scales, tables, and percentages. While the second method is cluster analysis with the help of SPSS 22 to classify slums based on the similarity of slum characteristics. Based on the analysis carried out, conclusions can be drawn regarding the slum clustering in Siak Regency, the clustering of slum settlements in Siak Regency was formed due to various characteristics that emerged in the slum areas of Siak Regency, among others: occupant characteristics, residential characteristics, infrastructure features. And the results of the analysis found 3 clusters of slums. Cluster 1 consists of a slum area in Kandis District and Minas District. Cluster 2 consists of the slum area of Tualang District, Upper Pipe Region. Cluster 3 consists of the slum area of Tualang Subdistrict Pipeline and Rear Market Area. Cluster 1 has the best infrastructure availability and physical condition among the three clusters. Cluster 3 has the worst infrastructure availability