Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statisti...Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.展开更多
The test selection and optimization （TSO） can improve the abilities of fault diagnosis, prognosis and health-state evalua- tion for prognostics and health management （PHM） systems. Traditionally, TSO mainly focuse...The test selection and optimization （TSO） can improve the abilities of fault diagnosis, prognosis and health-state evalua- tion for prognostics and health management （PHM） systems. Traditionally, TSO mainly focuses on fault detection and isolation, but they cannot provide an effective guide for the design for testability （DFT） to improve the PHM performance level. To solve the problem, a model of TSO for PHM systems is proposed. Firstly, through integrating the characteristics of fault severity and propa- gation time, and analyzing the test timing and sensitivity, a testability model based on failure evolution mechanism model （FEMM） for PHM systems is built up. This model describes the fault evolution- test dependency using the fault-symptom parameter matrix and symptom parameter-test matrix. Secondly, a novel method of in- herent testability analysis for PHM systems is developed based on the above information. Having completed the analysis, a TSO model, whose objective is to maximize fault trackability and mini- mize the test cost, is proposed through inherent testability analysis results, and an adaptive simulated annealing genetic algorithm （ASAGA） is introduced to solve the TSO problem. Finally, a case of a centrifugal pump system is used to verify the feasibility and effectiveness of the proposed models and methods. The results show that the proposed technology is important for PHM systems to select and optimize the test set in order to improve their performance level.展开更多
为了提高用于更新代理模型的解集的多样性和收敛性以提高代理模型准确度,提出一种基于行列式点过程(determinantal point process,DPP)的代理模型辅助多目标进化算法(surrogate-assisted evolutionary algorithm,SAEA)。首先,提出一种...为了提高用于更新代理模型的解集的多样性和收敛性以提高代理模型准确度,提出一种基于行列式点过程(determinantal point process,DPP)的代理模型辅助多目标进化算法(surrogate-assisted evolutionary algorithm,SAEA)。首先,提出一种基于行列式点过程的模型管理方法,从非支配解集基于行列式点过程选取子集并用真实目标函数评估,再从所有经真实目标函数评估的解中选取子集用于更新代理模型。另一方面,提出一种基于自适应行列式点过程的环境选择方法,在进化过程的早期侧重于提高种群的收敛性,在进化过程的后期侧重于提高种群的多样性。最后,基于DTLZ、WFG、MAF测试问题验证算法的有效性。将所提算法与K-RVEA、KTA2、CSEA等常用算法进行比较,使用IGD+指标进行评估。实验结果显示所提出的算法能得到更优的解集,从而证明了其高计算代价多目标优化问题上的有效性。展开更多
文摘Suppliers＇ selection in supply chain management （SCM） has attracted considerable research interests in recent years. Recent literatures show that neural networks achieve better performance than traditional statistical methods. However, neural networks have inherent drawbacks, such as local optimization solution, lack generalization, and uncontrolled convergence. A relatively new machine learning technique, support vector machine （SVM）, which overcomes the drawbacks of neural networks, is introduced to provide a model with better explanatory power to select ideal supplier partners. Meanwhile, in practice, the suppliers＇ samples are very insufficient. SVMs are adaptive to deal with small samples＇ training and testing. The prediction accuracies for BPNN and SVM methods are compared to choose the appreciating suppliers. The actual examples illustrate that SVM methods are superior to BPNN.
基金supported by the National Natural Science Foundation of China(51175502)
文摘The test selection and optimization （TSO） can improve the abilities of fault diagnosis, prognosis and health-state evalua- tion for prognostics and health management （PHM） systems. Traditionally, TSO mainly focuses on fault detection and isolation, but they cannot provide an effective guide for the design for testability （DFT） to improve the PHM performance level. To solve the problem, a model of TSO for PHM systems is proposed. Firstly, through integrating the characteristics of fault severity and propa- gation time, and analyzing the test timing and sensitivity, a testability model based on failure evolution mechanism model （FEMM） for PHM systems is built up. This model describes the fault evolution- test dependency using the fault-symptom parameter matrix and symptom parameter-test matrix. Secondly, a novel method of in- herent testability analysis for PHM systems is developed based on the above information. Having completed the analysis, a TSO model, whose objective is to maximize fault trackability and mini- mize the test cost, is proposed through inherent testability analysis results, and an adaptive simulated annealing genetic algorithm （ASAGA） is introduced to solve the TSO problem. Finally, a case of a centrifugal pump system is used to verify the feasibility and effectiveness of the proposed models and methods. The results show that the proposed technology is important for PHM systems to select and optimize the test set in order to improve their performance level.
文摘为了提高用于更新代理模型的解集的多样性和收敛性以提高代理模型准确度,提出一种基于行列式点过程(determinantal point process,DPP)的代理模型辅助多目标进化算法(surrogate-assisted evolutionary algorithm,SAEA)。首先,提出一种基于行列式点过程的模型管理方法,从非支配解集基于行列式点过程选取子集并用真实目标函数评估,再从所有经真实目标函数评估的解中选取子集用于更新代理模型。另一方面,提出一种基于自适应行列式点过程的环境选择方法,在进化过程的早期侧重于提高种群的收敛性,在进化过程的后期侧重于提高种群的多样性。最后,基于DTLZ、WFG、MAF测试问题验证算法的有效性。将所提算法与K-RVEA、KTA2、CSEA等常用算法进行比较,使用IGD+指标进行评估。实验结果显示所提出的算法能得到更优的解集,从而证明了其高计算代价多目标优化问题上的有效性。
