Abstract: Zinc ingots are the main raw material for the production of galvanized sheets and its consumption may fluctuate greatly due to the contract orders and product structure, which further results in fluctuating demand. Material demand often reflects the characteristics of small sample size and large variation range, whose non-stationarity and non-linearity make the demand forecasting more difficult. Meanwhile, the inaccuracy of demand forecasting will be gradually amplified in the information transmission of the supply chain, which will inevitably affect the material procurement plan and inventory management. Therefore, the accurate material demand forecasting has important practical significance for the optimization of raw material procurement and the production management scheduling of iron and steel enterprises. In order to improve the prediction accuracy of zinc ingot demand for galvanized sheet production, this paper proposes a zinc consumption prediction modeling method based on Support Vector Regression (SVR) optimized by Improved Grey Wolf Optimization (IGWO). Aiming at the shortcomings of fast convergence and premature maturity of traditional gray wolf algorithm, the chaotic Tent mapping strategy is firstly adopted to initialize the population so as to enhance the diversity and distribution uniformity of the initial population. An adaptive adjustment strategy of control parameters is introduced to balance the search ability and development ability of the algorithm. Finally, the differential evolution is integrated in the location update process to reduce the possibility of false convergence of the algorithm. For the improved gray wolf algorithm, a simulation experiment is made via a typical benchmark test function, whose result verify the superiority of the improved algorithm in comprehensive performance. Furthermore, based on the actual production data of a unit in a steel plant, the zinc ingot consumption is modeled and predicted, and the parameters of SVR is optimized via the IGWO algorithm. The experimental results show that IGWO-SVR has higher prediction accuracy, better stability and better generalization ability.