中文摘要：

为解决卷烟生产计划优化问题，根据卷烟生产工艺及产品特点，基于递阶生产计划（HierarchicalProductionPlanning，HPP）方法设计了卷烟生产计划的分层优化框架。该框架包括3层，最上层为总生产计划层，确定较长周期内产品族的生产安排；中间层为卷包分组计划层，确定卷包分组中各牌号的中长周期计划；最下层为制丝批次计划层，确定制丝生产批次投料顺序及时间。以总生产计划层为例，建立数学模型并利用动态规划方法求解，采用实际生产数据对该模型及求解方法的有效性进行仿真验证，结果表明：仿真数据所表示的平均库存水平和总生产成本比实际生产数据分别降低60％和30％以上，所建立的总生产计划模型优化效果显著。该方法为实现卷烟生产的成本最优化提供了参考。

英文摘要：

In order to optimize cigarette production planning, a 3-level optimization framework was designed based on Hierarchical Production Planning （HPP） method according to the production technology and characteristics of cigarette products. The framework is composed of three levels： aggregate production planning （top level）, it schedules the production of product family in a long term, making ＆ packing planning （middle level）, it determines the medium to long term plan of each cigarette brand, and batch planning of primary processing （bottom level）, it assigns the raw material input sequence and time of each batch in primary processing. Taking the aggregate production planning level as an example, a mathematical model was established, and it was solved with the dynamic programming method. The model and the solution were simulated and validated with actual production data. The results showed that comparing with the actual production data, the simulated average inventory level and total production cost reduced by more than 60% and 30% respectively, which indicated the significant effect of the established model. This method provides a reference for cost optimization in cigarette production .