Random-forest-based real-time contrasts control chart using adaptive breakpoints with symbolic aggregate approximation

作者：

Highlights：

• We propose a random-forest-based RTC chart with a data-representation method.

• Adaptive breakpoints with symbolic aggregate approximation improve the RTC chart.

• The represented data have categorical values, preserving information of the raw data.

• The proposed RTC control chart provides real-time detection and cause analysis.

• Experiments show that the proposed method outperforms the original RTC chart.

摘要

•We propose a random-forest-based RTC chart with a data-representation method.•Adaptive breakpoints with symbolic aggregate approximation improve the RTC chart.•The represented data have categorical values, preserving information of the raw data.•The proposed RTC control chart provides real-time detection and cause analysis.•Experiments show that the proposed method outperforms the original RTC chart.

论文关键词：Real-time contrasts (RTC),Control chart,Process monitoring,Symbolic aggregate approximation (SAX),Adaptive breakpoints-SAX (ABP-SAX)

论文评审过程：Received 17 September 2018, Revised 23 March 2020, Accepted 24 March 2020, Available online 12 May 2020, Version of Record 23 May 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113407