Monitoring of emulsion properties is important in many applications, since aged and broken emulsions perform worse and may affect product quality. Emulsion quality monitoring is a key issue in metalworking fluids (MWF), which are widely used in machining processes, since conventional control methods are not accurate. The present study reports results of the application of a UV-Vis-NIR spectroscopic sensor, coupled with a neural network model, as a new method of monitoring MWF emulsion destabilization, which is related to changes in its droplet size distribution. The technique has shown good accuracy in rebuilding the droplet size distribution of emulsions from spectroscopic measurements, indicating its feasibility to monitor MWF emulsion stability under operating conditions in industrial installations.