A surrogate-assisted particle swarm optimization framework for the integrated economic-statistical design of ANN-based profile monitoring
In the competitive landscape of Industry 5.0, the integrated optimization of Statistical Process Control (SPC), Maintenance Policies (MP), and Economic Production Quantity (EPQ) remains a critical challenge. The adoption of Artificial Intelligence (AI) models in a holistic economic-statistical frame...
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Article (Journal) |
| Langue: | anglais |
| Publié: |
October 2026
|
| In: |
Computers & industrial engineering
Year: 2026, Volume: 220, Pages: 1-16 |
| ISSN: | 1879-0550 |
| DOI: | 10.1016/j.cie.2026.112237 |
| Accès en ligne: | Verlag, kostenfrei, Volltext: https://doi.org/10.1016/j.cie.2026.112237 Verlag, kostenfrei, Volltext: https://www.sciencedirect.com/science/article/pii/S0360835226004389 
|
| Notes sur l'auteur: | Mohsen Shojaee, Arne Johannssen, Samrad Jafarian-Namin, Ali Yeganeh, Kim Phuc Tran, Nataliya Chukhrova |
| Résumé: | In the competitive landscape of Industry 5.0, the integrated optimization of Statistical Process Control (SPC), Maintenance Policies (MP), and Economic Production Quantity (EPQ) remains a critical challenge. The adoption of Artificial Intelligence (AI) models in a holistic economic-statistical framework is often hindered by a high computational burden. This study introduces a computationally efficient framework for the optimal design of Artificial Neural Network (ANN)-based control charts for linear profile monitoring. The framework synergizes a sensitive ANN-based control chart with a comprehensive 10-scenario cost model incorporating Delayed Monitoring (DM) policies. To overcome computational complexity, a high-fidelity ANN-based surrogate model, exhibiting a strong correlation with the true objective function and consistent performance across data subsets, is developed to efficiently guide a Particle Swarm Optimization (PSO) algorithm. Numerical results show that the proposed approach reduces optimization time by more than 95% while consistently outperforming a traditional statistical baseline. Economic improvements of up to 32.31% are achieved across various process scenarios. Additionally, the method demonstrates superior sensitivity to variance shifts, significantly reducing out-of-control detection delays. Thus, the framework enables a scalable and computationally tractable solution for the integrated design of intelligent quality control systems, effectively linking advanced monitoring performance with economic efficiency. |
|---|---|
| Description: | Online veröffentlicht: 3. Juli 2026, Artikelversion: 24. Juli 2026 Gesehen am 24.08.2026 |
| Description matérielle: | Online Resource |
| ISSN: | 1879-0550 |
| DOI: | 10.1016/j.cie.2026.112237 |