To address the issues of heavier crude oil and stringent product quality regulation，this paper presents a refinery planning model that integrates physical properties tracking and quality constraints.The proposed model selected crude oil’s key property of true boiling point curve as classification criteria and applied a K-means clustering to it，introduced swing cut to achieve a higher yield of valuable fractions，tracked and restricted petroleum products’ sulfur content as well as other key properties，and later proposed a universal solving strategy to deal with highly nonlinear quality constraints.The model was illustrated by representative case studies，the results showed that refinery planning optimization combined with quality tracking and constraints could satisfy product quality requirement while achieving a maximum refinery profit.