摘要
为实现供应链风险等级的高精度检测,基于SVM的参数设置对SVM的性能的影响,提出一种基于沙丘猫群算法(SCSO)优化SVM的供应链风险等级检测方法。首先,通过层次分析法建立供应链风险等级评价指标体系;之后,由于SVM的参数设置会影响到SVM的性能,利用SCSO算法对SVM的参数进行了优化,并给出了一种新的基于SCSO-SVM的供应链风险识别算法。与单独的SVM模型相比,SCSO-SVM的供应链风险检测的准确率分别提高了3.06、7.04个百分点,从而说明SCSO-SVM可以有效提高供应链风险检测的精度。
In order to realize the high-precision detection of supply chain risk level, based on the influence of SVM parameter setting on the performance of SVM,a supply chain risk level detection method based on SVMoptimized by sand cat swarm optimization(SCSO) is proposed.Firstly, the evaluation index system of supply chain risk grade is established by analytic hierarchy process;Then, because the parameter setting of SVM will affect the performance of SVM,SCSO algorithm is used to optimize the parameters of SVM,and a new supply chain risk identification algorithm based on SCSO-SVM is given.Compared with the single SVM model, the accuracy of supply chain risk detection of SCSO-SVM is improved by 3.06% and 7.04% respectively, which shows that SCSO-SVM can effectively improve the accuracy of supply chain risk detection.
作者
王宏刚
王一蓉
于宙
李君婷
孙妮
WANG Honggang;WANG Yirong;YU Zhou;LI Junting;SUN Ni(Big Data Center,State Grid Corporation of China,Beijing 100032,China)
出处
《粘接》
CAS
2023年第2期193-196,共4页
Adhesion
关键词
支持向量机
沙丘猫群算法
供应链
风险等级
support vector machine
sand cat swarm optimization algorithm
supply chain
risk level
