GA-LMBP Algorithm for Supply Chain Performance Evaluation in the Big Data Environment

10.3966/199115992017102805012

Pan Liu；Feng-Jie Zhang

big data ； genetic algorithm ； levenberg marquardt back propagation ； performance evaluation ； supply chain

電腦學刊

28卷5期（2017 / 10 / 01）

132 - 146

英文

An inefficient supply chain (SC) will lead to the resources waste. Big Data as a resource plays a vital role in improving SC performance. Therefore, to evaluate the effectiveness of SC and react the effects of bg data, to find a new evaluation method for SC performance evaluation is important. Firstly, in the traditional environment and Big Data environment, the previous performance evaluation indicator systems and algorithms were reviewed and discussed. Based on this, a five dimensional balanced scorecard was improved and proposed. In the improved five dimensional balanced scorecard, the Big Data usage indicators contained the capacity of gaining value and data leakage degree were proposed. Meanwhile, a method based on levenberg marquardt back propagation neural network algorithm and genetic algorithm was used for SC performance evaluation. Then, based on the practical data of company F, a case study was executed. Results shows that the method proposed has a high convergence speed and a precise prediction ability. The effectiveness and reliability of the model is confirmed. By comparing with the normal back propagation neural network algorithm, results indicates that the model proposed has a higher effectiveness and credibility. This method provides a suitable indicator system and algorithm for enterprises to implement SC performance evaluation in the Big Data environment. In theory, it is a new development of SC performance evaluation theory system and make up for the theory gap on SC performance evaluation. In practically, the method proposed has a theoretical guidance significance for enterprise to implement performance evaluation.