Scholarlink

<p>This study proposes a Supply Chain Operations Reference (SCOR®) based performance prediction model for the Make-to-Order job shop facility. The model uses Artificial Neural Network, which is fed with the real data collected from automotive job shop to predict cost and customer response using feed forward back error propagation learning algorithm with nonlinear activation function. The model was implemented using the MATLAB program and the correlation coefficient results demonstrated a high positive correlation between the expected and projected performance values, which supports SCOR® level 1 metrics for all ANN models. The average percentage error and the percentage standard deviation of the best cost model are found to be 0.75 and 1.28 respectively. Similarly, for the response model, they are found to be 0.13 and 0.25 respectively. These results highlight the quality of the developed model and its expected positive impact for improving supply chain performance.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Supply chain performance prediction model for make-to-order system using artificial neural network

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Sujan Piya,
  • Mahmoud Mokhtar

摘要

This study proposes a Supply Chain Operations Reference (SCOR®) based performance prediction model for the Make-to-Order job shop facility. The model uses Artificial Neural Network, which is fed with the real data collected from automotive job shop to predict cost and customer response using feed forward back error propagation learning algorithm with nonlinear activation function. The model was implemented using the MATLAB program and the correlation coefficient results demonstrated a high positive correlation between the expected and projected performance values, which supports SCOR® level 1 metrics for all ANN models. The average percentage error and the percentage standard deviation of the best cost model are found to be 0.75 and 1.28 respectively. Similarly, for the response model, they are found to be 0.13 and 0.25 respectively. These results highlight the quality of the developed model and its expected positive impact for improving supply chain performance.