Analisis Multikolinearitas, Distribusi Normal, Serta Korelasi Dan Regresi Pada Phosphogypsum Dengan Menggunakan Aplikasi Minitab

Abstract

Phosphogypsum is a chemical compound formed from the combination of gypsum (calcium sulfate) and phosphorus. This research aims to decompose and organize data from phosphogypsum to look for multicollinearity, normal distribution, as well as correlation and regression. In the multicollinearity results, it can be seen that the VIF value of C-H2O with CaSO4.2H2O is 559.37 and 562.71 and the VIF value of P2O5 tot with P2O5 WS is 10.82 and 10.81. In the normal distribution results, the CaO P-Value value was >0.150 and the Mesh P-Value value was 0.050. In the regression correlation results, in the final step, the R-square value for the morning shift P2O5 WS variable with the morning shift Seal Tank H2SO4 was 4.35%. It can be concluded, firstly, that the variables C-H2O with CaSO4.2H2O and P2O5 tot with P2O5 WS were detected as multicollinearity because they had VIF values >10. Second, the CaO and Mesh variables were detected to be normally distributed because they had a P-Value > 0.05. Third, the variable P2O5 WS morning shift with Seal Tank H2SO4 morning shift was detected to have a strong regression correlation because the P-Value was <0.05 and the regression value was greater than the other variables.