A Study of Compound Failure Rate Control Schemes for Monitoring IC Wafers Quality and Lifetime Reliability  
Author Yung-Yu Yang


Co-Author(s) Yun-Shiow Chen


Abstract The occurrence of defects clustered in the wafer of integrated circuit (IC), generally, violates the independency of the Poisson distribution and might misjudge the reliability and lifespan of wafers. Furthermore, the clustered defects and reliability tend to change over time which can increase the malfunction rate of the wafers. Disregarding the effect of clustered defects on an IC wafer’s reliability would result in incorrect estimation of the rates of Type I error and Type II error, and would trigger error signals on the quality control charts. We apply the EWMA control chart to discriminate subtle variations in the failure rate of the IC wafer and improve quality of the product as well. We construct a fault model for defects cluster in the early-life failure period based on a compound Poisson distribution for the cases of non-repairable IC wafers, and then consolidate the useful life period fault model and wear-out period fault model to form a composite failure rate for the IC wafer life cycle. In this research, we establish and propose a composite failure rate for the lifetime of IC wafers. Simulation results show that our model can work efficiently and accurately in studying the reliability of IC wafers. Furthermore, the EWMA control chart still works well in reflecting subtle change of the composite failure rate when the early-life failure time period extends to the lifetime of wafers.


Keywords Compound failure EWMA control chart, average run length, Markov chain, compound failure rate
    Article #:  1973
Proceedings of the 19th ISSAT International Conference on Reliability and Quality in Design
August 5-7, 2013 - Honolulu, Hawaii, U.S.A.