An Intelligent Factory Automation System With Multivariate Time Series Algorithm for Chip Probing Process

Abstract

Chip-probing is the key process for IC manufacturing to its ensure quality. As the number of tests increases, the test quality and the test yield will be affected because the needles on the probe card of the tester will be contaminated by external objects or worn out. Whether a needle polish of the probe card is required can be determined through real-time monitoring on various detection indicators such as resistivity and yield. However, both resistivity and yield are lagging indicators, and excessively frequent needle polishes will increase the processing time and reduce the test throughput. The so-called Intelligent Factory Automation (iFA) system platform, realized by integrating several intelligent services including Intelligent Predictive Maintenance (IPM), was proposed to accomplish the goal of Zero-Defect Manufacturing. However, the current remaining useful life (RUL) prediction algorithm in IPM is a univariate time series prediction. The RUL prediction may not be accurate enough if only one variable is adopted to describe the dynamic changes of the time series. A supervisory architecture for chip probing process based on iFA is proposed in this letter. The Multivariate Version of Time Series Prediction (TSP <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MVA</sub> ) in this architecture can use the vector autoregression model to improve the accuracy of RUL prediction. Experimental results reveal that the proposed supervisory framework with TSP <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MVA</sub> can not only monitor the tester's health status efficiently but also improve the accuracy of needle's RUL prediction by extracting multiple features.