Performance Measurement for SCM Based on Balanced Score Card and Self-Adaptive RBF Neural Network

According to the four perspective of BSC thinking, this article has selected new SCPM index system, which almost reflected multiple requests that meet the needs of customer and enterprises management. Meanwhile, based on typical and uniformity training samples selected by scientific uniform design method(UDM), the corresponding self-adaptive RBFNN model has been established. The experiments show the results between self-adaptive RBFNN evaluation and experts fuzzy comprehensive evaluation(FCE) are very close, the generalization of self-adaptive RBFNN with UDM is more better than that of the general RBFNN with Monte-Carlo method. The designed evaluation method realizes non-linear approaching ability of evaluation, meantime conquers the capability limitation of general RBFNN and BP neural network, and avoids the subjectivity and uncertainty of traditional FCE.