IDENTIFICATION OF REGIONAL TYPOLOGY IN THE PERSPECTIVE OF FOOD SECURITY IN BANYUMAS REGENCY. 2019. WANGSIT JUNIAWAN (Supervised by WIDODO & SRIYADI). The purpose of this research was analyze the status of sub-district food security based of food security indicators and grouping sub-district based on food security in Banyumas Regency. Location determination is done deliberately with consideration of low food productivity and high percentage of poverty in Banyumas Regency. To answer the fist research purpose used food security indicators FSVA as many as 6 indicators and to answer second researh puprose used cluster analylisis with the K-Means method. In the normative consumption ratio indicator, 1 sub-district is very vulnerable, vulnerable, quite resistant, and 2 resistant, 22 districts are very resistant. The indicator of the ratio of the number of poor family heads to 1 sub-district is very vulnerable and rather vulnerable, 8 resistant sub-districts, and 17 districts are very resistant. Indicators of the number of unemployed 1 sub-districts are almost vulnerable,1 subdistict resistant and 25 sub-districts are very resistant. Indicator of the number of malnourished toddlers 3 sub-districts are very vulnerable, vulnerable and quite resistant, and 7 sub-districts are rather vulnerable and very resistant, and 4 sub-districts are resistant. The indicator of the number of infant deaths in 2 sub-districts is very prone and vulnerable, 3 sub-districts are rather vulnerable, 5 sub-districts are quite resistant, 8 subdistricts are resistant, and 7 sub-districts are very resistant. Indicators of illiterate population 3 sub-districts are very vulnerable, 1 sub-district is vulnerable, 11 sub-districts are quite resistant, and 6 sub-districts are resistant and very resistant. Composite index 5 sub-districts are very resistant, 16 resistant sub-districts, and sub-districts hold a number of 6. The typology of the region is divided into 3 clusters, cluster I is the establishment of normative consumption ratio indicators with members of 4 sub-districts, cluster II is the indicator of the number of poor family heads with members of 2 subdistricts, and cluster III is the indicator of the number of infant deaths with 21 subdistricts.