Abstract: To address the structural imbalance in the supply of high-quality tobacco leaves and the high coupling between inventory and demand, and to enhance the digital design and maintenance capabilities of cigarette formulas, using a dataset of 482 types of inventory tobacco leaves, 23 cigarette brands, and a 3-year production plan, a quarterly-level rolling calculation process was constructed. Additionally, by integrating near-infrared spectroscopy technology, the sensory experience of formula designers, and computer-aided programming, a two-level similar substitution full-chain cigarette formula maintenance method was established, which combines the near-infrared-similar algorithm (SSD-2D-CORR) and the formula designer's experience-formula similarity algorithm (PLSR-AHP-PCA-MD). The results showed that: 1)Taking the TB2 brand cigarettes as an example for formula substitution, the differences in various quality indicators between the five new formulas and the original formula are all relatively small, meeting the requirements of the original formula. 2)Under the overall substitution calculation method, the independent control limits of various concerned indicators are relatively stable, significantly better than the individual substitution calculation method; through the comprehensive control limits, the independent control limits of various concerned indicators can be represented as a whole, which can reduce the difficulty for formulae in maintaining the stability of cigarette quality during the maintenance process. 3)This method can achieve automatic calculation of formula maintenance in the scenarios of full inventory, full brand, long cycle, and high coupling. It retains the global information of the spectrum and integrates the brand-oriented experience of the formulae. This research provides an effective solution for digital design of tobacco leaf procurement, inventory configuration, product maintenance, and cost control.