Cleaning processes in food production are often designed for the worst-case contamination to meet the required hygiene standards. They are consuming significant amounts of resources, which could be reduced using appropriate measures for assessing cleanliness. We have developed a fiber-optical sensor which provides a real time assessment of cleanliness inside closed vessels. It irradiates fouling deposits with violet light and records the fluorescence emission. The sensor tip can be integrated into any pipeline without perturbating the fluidic system. During production, fouling deposits accumulate similarly on top of the sensor as well as on the surrounding surfaces. We have validated the technique in a series of proof-of-principle experiments using model deposits such as whey protein concentrate simulated milk ultrafiltrate (WPC-SMUF), yoghurt, curd, mustard, apple juice and soda. The tests were performed in a small-scale industrial cleaning system where the determination of the equipment fouling condition by the sensor was validated by visual inspection. The sensor was able to detect the presence/absence of all model deposits during the cleaning process. Hence, the fiber-optical fluorescence sensor is a promising solution for monitoring the success of cleaning-in-place processes in food production.