联系我们

论文查询结果	返回搜索

论文编号：	14062
作者编号：	2120214096
上传时间：	2023/6/8 11:52:33
中文题目：	具有多能级的经济生产批量问题的改进遗传算法研究
英文题目：	An improved Genetic Algorithm for Economic Lot-Sizing problem with Multi-level Capacity
指导老师：	杨静蕾
中文关键字：	产能调整;经济生产批量;动态规划;遗传算法
英文关键字：	Dynamic Programming; Economic Lot-Sizing; Capacity adjustment; Genetic Algorithm
中文摘要：	制造业是我国经济的支柱产业。在工业4.0的背景下，制造业进行产业升级也是企业发展的必经之路。其中最优产能和产量计划是其中的重要一环。在需求多样化的新市场环境中，合理的产能和产量计划不仅能够降低企业的生产成本，还能够帮助企业实现对于顾客需求的快速响应。为了进一步提高企业的资源利用率，企业往往会采取多能级的生产线规划。产能不足时会导致缺货，而产能过剩则会造成浪费。因此企业应该准确的决策每个周期应该获取的产能能级，以及在当前产能能级下应该生产多少单位的产品。产能的不同又涉及到产能的调整，会产生产能调整成本和产能获得成本。基于以上情况，本文构造了以总成本最小化为目标的具有多能级的经济生产批量模型。 由于多能级的经济生产批量问题是NP-hard，本文设计了融合动态规划的遗传算法来求解具有多能级的经济生产批量模型，寻找使总成本最小的产能和产量计划。通过对模型的求解以及分析影响因素，得出了以下结论：（1）本文模型考虑了多能级和产能调整成本，相较于当前模型更接近真实场景。（2）本文设计的结合邻域搜索、动态规划的改进遗传算法比现有的经济生产批量问题算法的求解时间有所缩短。（3）产能、产量计划和总成本受单位生产成本的波动、产能调整成本和单位产能影响。单位生产成本波动越大总成本越小，产能调整成本越大其产能变动越小，而单位产能的数量也会影响到总成本。（4）当库存及缺货成本的和不变时，其比例对总成本影响不大。 根据以上结论，结合企业生产的影响因素，本文提出了相应的建议：（1）根据企业本身生产波动成本情况，来确定是采取原材料和成品库存囤积还是低库存的生产策略。（2）根据生产线调节难易程度，调整成本越高的企业越应该选择稳定生产的策略，而越低的应该越根据需求的波动调整产能。（3）企业不应该一味的追求精细化产能的管理，而应该综合考虑各种成本寻找一个合适的单位产能，以此来降低后期生产时的成本。在未来的研究中，可以考虑生产环节的不确定性因素，或者是产品的某些特性（比如易逝性），使得构建的模型更接近实际生产场景，进一步丰富该领域的研究内容。
英文摘要：	The manufacturing industry is a crucial component and serves as one of the main supports of China's economy. In the era of Industry 4.0, industrial upgrading has become an essential pathway towards the development of enterprises in the manufacturing industry. To achieve the transformation towards intelligent manufacturing, optimizing the planning of capacity and production is a crucial aspect for enterprises. In the current marketplace with diverse demands, strategic capacity and production planning can not only reduce production costs and improve profit margins for enterprises, but also enable them to respond quickly to customer demands. Enterprises frequently utilize multi-level capacity production line planning to optimize resource utilization efficiency. To avoid backlogging due to insufficient capacity or waste due to excess capacity, enterprises must make informed decisions about the number of production lines to open and the units of product to produce in each period. This requires careful consideration of all available production capacity levels, which may involve capacity adjustment and acquisition costs. Based on the above situation, this thesis constructs an economic lot-sizing model with multi-level capacity, with the goal of minimizing the total cost. As the Economic Lot-sizing Model with multi-level capacity is a NP-hard problem, this thesis proposes a genetic algorithm that incorporates dynamic programming to minimize the total cost by determining the optimal capacity and production plan. Through solving the model and analyzing the influencing factors, the following conclusions are drawn: (1) The model in this thesis considers multi-level production capacity and production capacity adjustment costs, which is more realistic than the current model. (2)By integrating neighborhood search and dynamic programming, the enhanced genetic algorithm introduced in this thesis significantly reduces the solving time when solving economic lot-sizing problems, as compared to existing algorithms. (3) The optimal capacity and production plan, as well as the total cost, are affected by the fluctuation of unit production cost, capacity adjustment cost, and unit production capacity. The total cost decreases as the fluctuation of unit production cost increases, while capacity stability increases with greater capacity adjustment costs. Additionally, the quantity of unit production capacity also has an impact on the total cost. (4) When the sum of inventory and backlogging costs remains unchanged, their proportion has little impact on the total cost. Based on the above conclusion and in combination with factors affecting enterprise production, this thesis proposes the corresponding suggestions: (1) Based on the fluctuation cost of production in the enterprise, decide whether to adopt a strategy of stocking raw materials and finished products or a low inventory production strategy. (2) Enterprises with high capacity adjustment costs should opt for a stable production strategy, while those with lower costs should adjust their capacity based on the fluctuation of demand. (3) Instead of blindly pursuing fine-grained production capacity management, enterprises should consider various costs holistically to determine the optimal unit production capacity that can reduce production costs in the later stages. In future research, uncertainty factors in the production process or certain product characteristics (such as perishability) can be considered to make the constructed model more applicable to actual production scenarios.
查看全文：	预览  下载(下载需要进行登录)