During the last 35 years, membranes have evolved from a laboratory tool to an industrial product with a wide variety of applications such as desalination of sea and brackish water and treatment of industrial effluents. Nowadays, a large number of research works are being done on making and development of ceramic membranes due to their better inherent stability at hard operating conditions in comparison with the polymeric membranes.agents. A number of experiments were conducted in order to verify the optimum value of gelling agents. These optimum values were found to be 5wt% and 7wt% for gelatin and CMC, respectively. Effects of sintering temperature, solid loading and median particle size of the alumina powder on the porosity of final casted part, were investigated. It was seen that the porosity decreased as the sintering temperature and the solid loading increased. On the other hand, at similar sintering temperature and solid loading conditions, powders with larger particle diameter resulted in higher porosity. Dip-coating method was used to modify the surface of manufactured supports. In this part of the work, a number of alumina supports with the same characteristics were produced and used for dip-coating step. Taguchi method was used to design the experiments and the quantitative effects of important parameters namely suspension concentration, dipping time, withdrawal speed, and the number of coated layers on separation properties of the membranes were investigated. Three levels were selected for the first three parameters and the last parameter was adjusted at two levels. Two objective functions of the permeated flux and rejection were defined to evaluate the performance of the membranes. These membranes were tested in separation of 5 gr/Lit PEG10000/water solutions. Analysis of variance (ANOVA) showed that the number of coated layers is the main effective parameter on the membrane flux and rejection. Using Taguchi analysis, optimum conditions for synthesis of the optimized membranes for each objective function were determined. These membranes were different in dip-coating conditions. The membrane coated twice by 20wt% suspension with dipping time of 20 second and withdrawal speed of 72 mm/min had the optimum rejection of 0.962. Also, the membrane with one coated layer coated by 20wt% suspension, dipped for 30 second with a withdrawal speed of 36 mm/min showed the maximum flux of 6.53*10 -6 m 3 /m 2 .s The rejection obtained by the most of the synthesized membranes were higher than 90% for filtration of low concentration PEG10000-water solution, which means these membranes had MWCO of 10000. Keyword: Gel Casting; Dip-Coating; Taguchi Design of Experiments; MWCO.