A web transfer process is composed of many drive rolls to control the tension and the speed at a desired value to secure the integrity of the operation. This research aims to verify our proposed method of automatic tuning of the controllers parameters using a neuro-fuzzy logic control (NFLC) for the web tension decentralized control system. This paper shows the feasibility of the NFLC; during the operation, the parameters of the fuzzy controllers are adjusted by using the back-propagation method. The membership functions and the output of the fuzzy interference are adaptively tuned based on the output errors, and those magnitudes become modest as learning proceeds.