The pipeline scheduling plan is an important operation basis for the multi-product pipeline to complete its transportation tasks. In addition to ensuring the delivery of products to the stations along the route on time and in the right quantity, it should also maintain the stability of the pipeline as much as possible and improve the overall service quality of the pipeline scheduling. At present, for the multi-product pipeline with centralized distribution process, the domestic and foreign studies on scheduling plan development have not been able to take the stability of pipeline operation into consideration. Moreover, some literature confirms that if the pipeline flow is only controlled by hydraulic constraints, it will cause frequent fluctuations in a certain range, which will weaken the safety of pipeline operation, resulting in the research results being detached from the engineering reality and rarely applied on the ground. In this paper, based on the existing research, the key parameters adapted to the centralized distribution process, such as flow fluctuation coefficient and upper limit of operation number, were introduced for the single-source multi-sink pipeline, and the optimization strategy was proposed based on the discrete-time expression method. Thereafter, constraints such as station delivery process, pipeline section operation flow rate, batch tracking, estimated delivery volume, delivery deviation, and flow rate deviation are considered, and the delivery deviation and operation flow fluctuation are minimized as the objective function. Combining with the field experience of manual preparation plan for changing the delivery flow rate of this station at the operation time node in adjacent stations, an automatic preparation model of pipeline scheduling plan is constructed to realize the rapid formulation of pipeline scheduling plan for multi-products considering the stable operation. The model was applied to an actual multi-product pipeline containing 9 stations in China, and the monthly scheduling plan of the pipeline could be solved accurately within 2 minutes; the scheduling plan completely considered the entire delivery process and one-time delivery requirements of the pipeline, and met the generality requirements; the solved results accounted for 90% of the normal flow fluctuation during operation, and the deviation of product demand was controlled within -3.23%~3.65%, which has strong engineering practicality. The research results have certain guiding significance for the multi-product pipeline operation enterprises adopting the centralized distribution process in formulating the scheduling pla