Analisis Sensitivitas Metode Vikor dan Topsis pada Indeks Pembangunan Manusia di Indonesia

Abstract

Sensitivity analysis is an important step in decision making, especially in determining how sensitive the ranking results are to changes in the weights of the criteria used. This research aims to analyze the sensitivity of the VIKOR and TOPSIS methods in ranking the Human Development Index (HDI) in Indonesia. This research also compares the sensitivity and similarity of rankings of the VIKOR and TOPSIS methods in ranking the Human Development Index (HDI) in Indonesia in 2022 with 4 criteria aspects, namely life expectancy, expected length of schooling, average length of schooling, and real expenditure per capita. The research results show that both methods produce a significant ranking equal to the Central Statistics Agency (BPS) ranking. Sensitivity analysis with changes in weight criteria of 1%, 25%, and 50% shows that the VIKOR method is more sensitive to changes in weight than the TOPSIS method. At a weight change of 1%, VIKOR experienced changes while TOPSIS showed high stability without changes in ranking. However, at weight changes of 25% and 50%, both methods showed high sensitivity. Additional testing of the TOPSIS method with 3% and 7% weight changes shows that TOPSIS remains stable at a 3% weight change for the real per capita expenditure criterion, but begins to show ranking changes at a 7% weight change for all criteria. So the VIKOR method is more suitable for ranking HDI in Indonesia in various situations of changing weights, in line with the changing dynamics of HDI to improve human development in Indonesia. Meanwhile, the TOPSIS method is more suitable for situations that require stability in small weight changes but remains adaptive to larger changes.