Aggregated parameter update schemes for monitoring binary profiles

Area of Science:

• Statistical Process Control
• Quality Engineering
• Data Analytics

Background:

• Profile monitoring is crucial for statistical process control, but traditional methods struggle with massive datasets.
• Existing schemes require extensive historical data for parameter estimation, posing computational challenges.
• Modern processes generate large sample sizes, necessitating efficient parameter update strategies.

Purpose of the Study:

• To develop a novel recursive update strategy for profile monitoring of massive datasets.
• To design a self-starting control chart that efficiently handles large-scale process data.
• To improve parameter estimation accuracy and process monitoring effectiveness in complex industrial settings.

Main Methods:

• A recursive update strategy based on the aggregated estimation equation (AEE) was developed.
• A self-starting control chart was designed using the AEE for binary profile monitoring.
• Numerical simulations and a real-data example were used for validation.

Main Results:

• The proposed AEE-based recursive update strategy demonstrated superior performance in parameter estimation.
• The novel self-starting control chart showed enhanced effectiveness in process monitoring.
• Asymptotic properties of the monitoring statistic were derived, confirming theoretical soundness.

Conclusions:

• The AEE-based recursive update strategy offers an economical and efficient solution for profile monitoring of massive datasets.
• The developed self-starting control chart significantly improves upon traditional methods for complex statistical processes.
• The findings have practical implications for quality control in industries dealing with large-scale data.