Editing Design of Statistical Quality Control Procedures Using Genetic Algorithms

Title: Design of Statistical Quality Control Procedures Using Genetic Algorithms

Abstract: This research aimed to explore the application of genetic algorithms (GAs) in designing statistical quality control (QC) procedures for analytical processes. The study developed an interactive GA-based computer program that designs a novel near-optimal QC procedure given an analytical process. The program uses the deterministic crowding algorithm and can detect critical random and systematic errors with stated probabilities, while minimizing the probability of false rejection. An illustrative application suggests that the program has the potential to design QC procedures significantly better than 45 alternative ones used in clinical laboratories.

Research Question: Can genetic algorithms be used to design optimal statistical quality control procedures for analytical processes?

Methodology: The study employed a computational approach using genetic algorithms. The algorithm was designed to search through large parameter spaces quickly, optimizing a QC procedure without requiring knowledge of the objective function. The program used the deterministic crowding algorithm and was applied to an illustrative case in clinical chemistry to demonstrate its potential.

Results: The study found that the GA-based program could design a novel near-optimal QC procedure, detecting critical random and systematic errors with stated probabilities and minimizing the probability of false rejection. The illustrative application suggested that the program had the potential to design QC procedures significantly better than 45 alternative ones used in clinical laboratories.

Implications: The research implies that genetic algorithms can be used to design optimal statistical quality control procedures for analytical processes. The interactive GA-based computer program developed in the study can optimize QC procedures without requiring knowledge of the objective function, making it a promising tool for improving the performance of analytical procedures in various fields.

Link to Article: https://arxiv.org/abs/0201024v2
Authors: 
arXiv ID: 0201024v2

[[Category:Computer Science]]
[[Category:Procedures]]
[[Category:Qc]]
[[Category:Program]]
[[Category:Design]]
[[Category:Genetic]]