The main objective of this thesis is to offer a practical solution to the problems encountered in color-critical industrial applications, caused by individual variability among observers with normal color vision, commonly referred to as observer metamerism. This work starts by conducting a comprehensive theoretical analysis on various aspects of the physiologically-based observer model proposed in 2006. In the context of color perception on modern narrow-band displays, the performance of this model in predicting average observer data was evaluated, and based on a nonlinear optimization, an improvement of the model was proposed. Next, several color-matching experiments were performed on two displays, confirming the effect of observer metamerism in display color matches. Then, based on a statistical analysis of a comprehensive visual dataset, eight colorimetric observer categories were identified. An experimental observer classification method was developed, and was implemented by means of a compact prototype, the Observer Calibrator. Visual experiments performed on the prototype proved that a small number of categories can be assigned to color-normal observers based on their color vision. Finally, an implementation of colorimetric observer categories in relevant industrial applications has been proposed, and nonlinear transformations that result in accurate color transformations between categories have been derived. It is hoped that the observer classification method, together with the compact and economical prototype, will be valuable contributions not only for industrial applications, but also for scientific research in the domains of color science and vision.