Feasibility Study on the Use of a Portable NIR Spectrometer and Multivariate Data Analysis to Discriminate and Quantify Adulteration in Fertilizer

Table 6

Optimum simultaneous quantification of fertilizers A and B adulteration levels.

Model	R^2	RMSECV	Bias	R^2	RMSEP	Bias	RPD
PLS	0.8133	0.2064	0.0102	0.7966	0.2139	0.0161	1.6589
iPLS	0.8920	0.1603	−0.0003	0.8824	0.1692	−0.0198	2.0972
Bi-PLS	0.9156	0.1426	0.0002	0.7068	0.2537	−0.0366	1.3987
Si-PLS	0.9460	0.1150	−0.0003	0.9295	0.1328	−0.0193	2.6720
GAPLS	0.8895	0.1620	−0.0002	0.8551	0.1864	−0.0135	1.9020
		Variables			Selected wavelength range (nm)
PLS		331			740–1070
iPLS		16			959–974
Bi-PLS		50			740–756, 808–824, 927–942
Si-PLS		61			740–755, 816–830, 936–950, 966–980
GAPLS		331			740–1070