Monitoring linear profiles using Artificial Neural Networks with run rules
作者：
Highlights：
• An ANN-based chart with novel run rules is proposed for monitoring linear profiles.
• A heuristic procedure is introduced for the design of the proposed chart.
• The proposed chart outperforms other competing methods in most of the cases.
• Adding run rules significantly improves the efficiency of ANN-based chart.
• The proposed run rules is more effective than traditional Western Electric rules.
摘要
•An ANN-based chart with novel run rules is proposed for monitoring linear profiles.•A heuristic procedure is introduced for the design of the proposed chart.•The proposed chart outperforms other competing methods in most of the cases.•Adding run rules significantly improves the efficiency of ANN-based chart.•The proposed run rules is more effective than traditional Western Electric rules.
论文关键词：Artificial Neural Networks (ANN),Average run length (ARL),Simple Linear Profile,Phase II,Profile diagnosis,Profile monitoring
论文评审过程：Received 9 April 2020, Revised 5 August 2020, Accepted 3 November 2020, Available online 10 November 2020, Version of Record 24 January 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114237
