Multi‐Objective Optimization for Short‐Term Scheduling Considering Crude Oil Mixing Requirements Based on Improved NSGA‐III

ABSTRACT This paper addresses the crude oil short‐term scheduling problem considering dynamic mixing requirements, aiming to more comprehensively reflect production realities and optimize costs. Based on the given refining schedule at the upper level, a novel multi‐objective optimization model is proposed to optimize five key objectives. To solve this model, a new GNSGA‐III algorithm is introduced, which incorporates an initialization strategy based on a good point set, along with enhanced crossover and mutation operators to improve population diversity. The results show that the GNSGA‐III algorithm outperforms the NSGA‐III algorithm, achieving an improvement of 11.6%–26.7% in optimization performance across multiple objectives. Furthermore, the overall scheduling cost is significantly reduced, demonstrating the effectiveness of the proposed approach. The study contributes to optimizing crude oil scheduling by incorporating mixing constraints and improving algorithm efficiency, providing a more cost‐effective solution for the practical scheduling problems of refineries.