Adaptive Fault-Tolerant Control in Automated Manufacturing Systems Using Hybrid AI-FIS Models

Abstract

Automated manufacturing environments demand high operational reliability and robust control strategies to ensure consistent product quality and uninterrupted production. However, unexpected faults in actuators, sensors, and communication systems can degrade performance or halt production entirely. This paper presents an adaptive fault-tolerant control framework based on a hybrid Artificial Intelligence–Fuzzy Inference System (AI-FIS) model that detects, isolates, and compensates faults in real-time. The hybrid model leverages rule-based fuzzy reasoning for interpretability and neural adaptation for dynamic learning. A multi-stage controller continuously monitors system states, estimates fault severity, and applies corrective actions without compromising stability. Experimental evaluations in automated assembly and precision machining systems demonstrate improvements in fault detection speed, system reliability, and control accuracy. The proposed approach offers a robust and flexible solution for maintaining high-quality automation performance under uncertain and fault-prone working conditions.

KEYWORDS: Fault-Tolerant Control, Fuzzy Inference System, Automation, Fault Diagnosis, Hybrid AI Models