A REVIEW ON: SELF-LEARNING FAULT PROGNOSTICS FRAMEWORK FOR PLC-INTEGRATED AUTOMATED STORAGE AND RETRIEVAL MECHANISMS

Abstract

Automatic Storage and Retrieval Systems (AS/RS) are widely used in modern automated warehouses to improve efficiency, accuracy, and productivity in material handling operations. These systems are typically controlled by Programmable Logic Controllers (PLC), which manage the movement of storage and retrieval mechanisms using sensor inputs and control logic. However, PLC-based AS/RS are prone to unexpected faults due to mechanical wear, sensor failures, communication errors, and control system issues, which can lead to system downtime and reduced operational efficiency. Traditional maintenance approaches mainly rely on periodic inspections or reactive maintenance and often fail to detect faults at an early stage. To address this problem, this study aims to develop a data-driven fault prediction framework using advanced machine learning techniques. Operational data from PLC sensors, actuators, and system logs will be collected and preprocessed through data cleaning, normalization, and feature extraction to create a reliable dataset. Machine learning algorithms such as Random Forest, Support Vector Machine (SVM), and Artificial Neural Networks (ANN) will be applied to analyze operational patterns and identify early indicators of potential system failures. The proposed framework will evaluate and compare the performance of different models in predicting faults and improving system reliability. The expected outcome of this research is the development of an intelligent predictive maintenance system capable of detecting faults in advance, reducing downtime, optimizing maintenance scheduling, and enhancing the overall reliability and efficiency of PLC-based AS/RS in automated warehouse environments.