Improving Statistical Process Control Using Artificial Neural Networks
Loading...
Date
2025-07-21
Authors
Zahra Maqboul
Amani Nassasra
Ehab Mashaqi
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Abstract
Today's manufacturing processes acutely pressurize all quality control systems, including the
stability in process conditions and product conformity. The performance of the statistical
process control will be further improved with machine learning methods, including ANN
applications; their modeling will contribute much more to the effective monitoring and
prediction of process variables. The research will study the process of ice cream production at
Al-Arz Ice Cream Company and utilize historical data to study key quality indicators such as
box weight, casting weight, chocolate temperature, and the weight of added chocolate.
The objective of this work was to propose a predictive model, that will optimize the quality
control process by joining SPC principles with ANNs. A dataset collected over four months
was analyzed, preprocessed, and then used to train multiple ANN models. These models have
been evaluated for performance and compared to traditional SPC methods. The developed
models showed more accuracy for the identification of complex patterns and prediction of
deviations from the quality standard.
ANN implementation in this project showed significant improvement in the effectiveness of
quality control, reducing the chances of error appearance and making production processes
stable. Data inconsistencies and model parameter optimizations are the major challenges. The
results showed that machine learning could offer new impulses for a completely changed
quality management approach; the early detection of variations of the processes, while realtime monitoring is enabled, can offer a scalable, adaptive, and effective solution for
contemporary industries.