基于可制造性特征的制造资源建模  

Manufacturing Resource Modeling Based on Manufacturability Characteristics

作　　者：赵昌龙[1] 杨俊宝 李明 赵钦祥 马洪楠 贾晓宇 ZHAO Changlong;YANG Junbao;LI Ming;ZHAO Qinxiang;MA Hongnan;JIA Xiaoyu(College of Mechanical and Vehicle Engineering,Changchun University,Changchun Jilin 130022,China)

机构地区：[1]长春大学机械与车辆工程学院,吉林长春130022

出　　处：《机床与液压》2023年第14期132-138,共7页Machine Tool & Hydraulics

基　　金：吉林省教育厅科技计划项目(JJKH20220589CY)。

摘　　要：可制造性评价是缩短开发周期、优化制造工艺、降低产品成本的有效途径。产品的可制造性取决于特定制造资源的加工能力。制造资源模型的开发是可制造性评价的基础。为了更好地利用制造资源信息,提出一种模糊C均值聚类算法与遗传算法相结合的混合算法,根据制造特征对制造资源进行分组,基于制造资源约束的Object-Oriented方法建立制造资源信息模型,设计可制造性评价框架。通过对32台加工设备进行划分,使用混合算法动态确定最优分组数目和该数目下的最优分组。结果表明:混合算法可靠有效,能够提高应用企业的整体绩效,增强决策的可行性,并有利于管理层做出更明智的决策。Manufacturability evaluation is an effective way to shorten development cycles,optimize manufacturing processes,and reduce product costs.The manufacturability of a product depends on the processing capability of a specific manufacturing resource.The development of manufacturing resource model is the basis of manufacturability evaluation.In order to make better use of manufacturing resource information,a hybrid algorithm combining fuzzy C-mean clustering algorithm and genetic algorithm was proposed to group manufacturing resources according to manufacturing characteristics,a manufacturing resource information model was established based on the Object-Oriented method with manufacturing resource constraints,a manufacturability evaluation framework was designed.The hybrid algorithm was used to dynamically determine the optimal number of groupings and the optimal grouping under the number by dividing 32 sets of processing machine.The results show that the hybrid algorithm is reliable and effective in improving the overall performance of the company in which this study is applied,enhancing the feasibility of decision making,and facilitating more informed decisions by management.

关 键 词：可制造性 制造资源模型 模糊C均值聚类算法 遗传算法 

分 类 号：TP156[自动化与计算机技术—控制理论与控制工程]