Information from the abstract
This study evaluates the performance of six generative artificial intelligence (AI) systems in solving Thai-language multiple-choice examinations in the subject of Programmable Logic Controllers (PLC), a core component of Industrial Electrical Engineering education. Six large language models (LLMs), including ChatGPT, Claude, DeepSeek, Gemini, Copilot, and Grok, were tested using fifteen sets of PLC examination questions. Statistical analysis was conducted using one-way ANOVA and two-sample t-Tests with Bonferroni correction to examine performance differences. In addition, effect size measures, including Eta-squared (η²) and Cohen‟s d, were calculated to assess the magnitude of the observed differences. The results show that ChatGPT achieved the highest mean score (77.27%), while DeepSeek followed closely (76.73%) and demonstrated the lowest standard deviation (±1.83%), indicating the most consistent performance across test sets. Claude also performed strongly (74.80%), whereas Gemini, Copilot, and Grok achieved similar mid-tier scores ranging from 72.40% to 72.73%. Although all LLMs achieved scores within the passing grade range, ANOVA confirmed statistically significant differences among systems (p-value is 0.0002). However, after applying the Bonferroni correction, only a subset of pairwise differences remained statistically significant, particularly between DeepSeek and several mid-tier LLMs, while the differences among ChatGPT, Claude, and DeepSeek were not statistically significant under the adjusted threshold. Effect size analysis further indicates that some of these differences represent meaningful practical variation in LLM performance. These findings indicate that contemporary LLMs demonstrate baseline comprehension of PLC concepts and can achieve passing-level performance in technical examinations conducted in a non-English language. The study contributes empirical evidence on AI performance in Thai-language technical assessments and highlights the potential role of generative AI as a complementary learning support tool in vocational and engineering education.
Why this record is monitored
This record has an Impact Signal of 81/100 based on recency, source, collaboration, and bibliographic signals. It prioritizes monitoring and is not a judgment of research quality.
Related topics: Artificial Intelligence in Healthcare and Education · AI in Service Interactions · Digital Transformation in Industry
Thai researcher and institutional participation
Amnaj Prajong · Therdpong Daengsi · Nakhon Sawan Rajabhat University · Rajamangala University of Technology Phra Nakhon
Data limitations
This page is a bibliographic record based on abstract-level information, not a full analysis or quality assessment. Verify the DOI and original article before citation.