Integrating AI And Data Analytics In Assessing The Learning Effectiveness Of A PLC Training Model

Abstract

With the rapid advancement of industrial automation, Programmable Logic Controllers (PLCs) have become essential components in modern manufacturing systems. To enhance practical training in PLC programming, a hands–on experimental model based on the Mitsubishi FX5U PLC was developed. This study evaluates the effectiveness of the training model through student feedback collected via structured survey questions, incorporating both quantitative and qualitative measures. A Likert–scale rating system was utilized to assess key aspects such as conceptual understanding, programming proficiency, and practical applicability. Additionally, qualitative responses were analyzed using AI–driven sentiment analysis and data analytics techniques, including lexicon–based (VADER) and transformer–based (BERT) models. Word Cloud visualizations were employed to extract key insights from open–ended responses. The study offers a detailed evaluation of the PLC training model, identifying both its effective components and aspects requiring enhancement. By leveraging AI–driven evaluation techniques, the research demonstrates how data–informed insights can enhance instructional quality and inform future improvements in technical education.