The proportioning of iron ore is the first step of the sintering process. It mixes different kinds of iron ores with coke, limestone, dolomite, and returned sinter to produce a raw mix for the production of qualified sinter. The chemical components and proportions of the raw materials determine the chemical and physical characteristics of the resulting sinter, and thus the quality of the sinter and the amount of SO2 emissions. The prices of the raw materials and their proportions determine the price of the sinter. In this study, an intelligent integrated optimization system (IIOS) was developed for the proportioning step, which contains two phases: the first and second proportionings. First, the sintering process was analyzed, and the requirements of the proportioning step were specified. Next, an IIOS with two levels (intelligent integrated optimization, basic automation) was built. In the intelligent integrated optimization level, an intelligent integrated optimizer (IIO) produces an optimal dosing scheme. The IIO has three parts: a cascade integrated quality-prediction model, the optimization of the first proportioning, and the optimization of the second proportioning. Computational intelligence methods predict the quality of sinter. Then, the predicted quality indices are fed back to the optimizations of the first and second proportionings to find feasible optimal dosing schemes. The IIOS was implemented in an iron and steel plant. Actual runs show that the system reduced production costs by 43.014 CNY/t and SO2 emissions by 0.001% on average. (C) 2013 Elsevier Ltd. All rights reserved.