In a hot-dip galvanizing process, the mass of zinc deposited and its distribution are controlled by the air pressure and effective distance from the air knife nozzle to the steel strip surface. In this paper, we investigate how modem techniques of modeling, estimation and control can be used to design an advanced air knife controller that improves the uniformity of the deposited mass of zinc on steel strip surfaces and reduces zinc consumption. An algorithm consisting of Kalman filters and knife position and air pressure controllers has been developed for the air knife. Simulation and on-line trail results confirm the feasibility and reliability of the proposed algorithm. These results also illustrate the unique and excellent performance of the algorithm, and suggest that the potential annual saving of zinc consumption is of the order of US$1M.