One of important quality parameters of white crystal cane sugar is its color, which is measured as the ICUMSA value referring to the standard method established by the International Commission for Uniform Methods of Sugar Analysis (ICUMSA). It is usually measured in a laboratory using a complex and lengthy chemical analysis method. To overcome this challenge, this research attempts to explore the potential use of multi-channel spectral sensors in the UV-Vis-NIR region as an alternative method to predict the ICUMSA value. The proposed portable device uses an AS7265X sensor as the main component. The spectra data of 60 cane sugar samples were collected using the proposed device followed by measurements of ICUMSA value in the laboratory using standard methods as reference. The prediction using partial least squares regression (PLSR) model achieved R2 = 0.896, RMSEC = 0.072%, RMSEP = 0.103%, CV = 26.087%, and RPD = 3.104. The multiple linear regression (MLR) model achieved R2 = 0.910, RMSEC = 0.067%, RMSEP = 0.111%, CV = 24.328%, and RPD = 3.328. The artificial neural network (ANN) model achieved R2 = 0.999, RMSEC = 0.004%, RMSEP = 0.037%, CV = 1.433% and RPD = 9.543. This result indicates that the developed PLSR, MLR, and ANN models can predict the ICUMSA value well with ANN as the best model. It also can be concluded that the proposed portable device can be an alternative for rapid analysis of ICUMSA value.