Zirconium nitride (ZrN) and titanium nitride (TiN) thin films were deposited on Si(1 0 0) substrates using hollow cathode discharge ion-plating (HCD-IP) and unbalanced magnetron (UBM) sputtering methods, respectively. Design of experiment (DOE) has long been recognized as a powerful method to optimize a complex process in industry. The purposes of present study were to verify the feasibility and reliability of the application of DOE method on PVD processes and optimize the processing parameters for the deposition process, in which the sensitive parameters that affected the film properties were also identified. For HCD-IP method, four parameters, including substrate temperature, substrate bias, Ar partial pressure, and N-2 partial pressure, were selected; for the UBM method, dc power, N-2 flow rate, specimen-target distance, and specimen height were chosen to be the operating parameters. After deposition, the thin film structure was characterized by X-ray diffraction (XRD), and high-resolution scanning electron microscopy (SEM). The hardness of the ZrN and TiN films was determined by nanoindentation. The resistivity of thin films was measured by a four-point probe. The roughness of the film was obtained by atomic force microscopy (AFM). The analysis of mean (ANOM) and analysis of variance (ANOVA) were conducted to assess the sensitive parameters and predict the optimum conditions. Based on the statistical analysis, the most sensitive parameters in the HCD-IP process were the substrate bias and the N-2 partial pressure and in the UBM process were the dc power and the specimen height. The optimum deposition conditions in each system were also acquired. (C) 2003 Elsevier Science B.V. All rights reserved.