Abstract: This study delves into the comprehensive optimization challenge of integrating production, operation, sales, and storage of the industrial chain when large-scale energy enterprises formulate their operational plans. To address the limitations of existing models, which typically cover only a single time period, ignore business conflicts, and treat transportation and inventory separately, this paper develops a multi-period inventory-transportation joint optimization model. The objective is to maximize the total profit and minimize the impact of business conflicts within the enterprise planning horizon. This model considers constraints such as inventory balance across different time periods and production-operation-sales activities, while business conflicts are also taken into account. Given the complexity of the model structure, numerous constraints, and substantial computation cost, this paper introduces a fast genetic algorithm with progressive enhancement of soft constraints. Compared to traditional genetic algorithms, this algorithm can accelerate computation by 5-10 times, significantly improving the efficiency of solving such models. Finally, the model and algorithm are applied to practical cases, while detailed analysis of these cases is conducted, verifying the reliability of the model and the efficiency of the algorithm. The model and algorithm proposed in this paper contribute to providing scientific, rational, efficient, and precise decision support for the optimal operations of the integrated industrial chain in large-scale energy enterprises.

Key words: inventory-transport joint optimization, multi-period, business conflicts, fast genetic algorithm