EVALUATING THE CAPABILITY OF THE GPT-5 LANGUAGE MODEL IN SOLVING REAL - WORLD OPTIMIZATION PROBLEMS IN GRADE 12 MATHEMATICS

EVALUATING THE CAPABILITY OF THE GPT-5 LANGUAGE MODEL IN SOLVING REAL - WORLD OPTIMIZATION PROBLEMS IN GRADE 12 MATHEMATICS

Do Ngoc Liem liemdongoc523@gmail.com Independent Researcher 98A, 8C road, Trung Son residential area, Binh Hung commune, Ho Chi Minh City, Vietnam
Summary: 
In the context of educational reform and the rapid development of artificial intelligence (AI), assessing the applicability of AI in Mathematics education has become increasingly essential. This study investigates the capability of GPT-5 in solving real-world optimization problems included in the grade 12 Mathematics (2018), which are often challenging for students. A dataset of 40 representative problems was constructed and solved using three different prompt configurations. The AI-generated solutions were evaluated based on a rubric with five criteria: mathematical modelling, analytical process, boundary analysis, practical interpretation, and presentation. Two Mathematics teachers independently graded the solutions, and inter rater reliability was verified using Cohen’s Kappa alongside statistical analyses. The results indicate that GPT-5 demonstrates strengths in mathematical modelling, boundary analysis, and structured presentation but reveals limitations in analytical rigor and practical interpretation. These findings provide scientific evidence on both the potential and the constraints of AI in Mathematics education. They also suggest that AI should be employed as a supportive tool under teacher guidance rather than as a replacement in developing students’ mathematical competencies.
Keywords: 
Artificial intelligence
GPT-5
optimization
real-world problems
mathematics education
grade 12 Mathematics.
Refers: 

[1] Awang, L. A., Yusop, F. D. & Danaee, M. (2025). Current practices and future direction of artificial intelligence in mathematics education: A systematic review. International Electronic Journal of Mathematics Education, 20(2), em0823. https:// doi.org/10.29333/iejme/16006

[2] Borah, P. (2024). A review of use of artificial intelligence in teaching and learning of mathematics. International Journal on Science and Technology, 15(4), 1–12. https://www.ijsat.org

[3] Bộ Giáo dục và Đào tạo. (2018a). Chương trình Giáo dục phổ thông môn Toán (Thông tư 32/2018/TT-BGDĐT). NXB Giáo dục Việt Nam.

[4] Bộ Giáo dục và Đào tạo. (2018b). Chương trình Giáo dục phổ thông - Chương trình tổng thể (Thông tư 32/2018/ TT-BGDĐT). NXB Giáo dục Việt Nam.

[5] Du, W., Cao, Y., Tang, M., Wang, F. & Wang, G. (2025). Factors influencing AI adoption by Chinese mathematics teachers in STEM education. Scientific Reports, 15, 20429. https://doi.org/10.1038/s41598 025-06476-x

[6] Hwang, G.-J. & Tu, Y.-F. (2021). Roles and research trends of artificial intelligence in mathematics education: A bibliometric mapping analysis and systematic review. Mathematics, 9(6), 584. https:// doi.org/10.3390/math9060584

[7] Lee, S.-G., Lee, J. H., Nielsen, Y. L. & Ko, H. K. (2021). Case study of an online course on introductory mathematics for artificial intelligence. The Electronic Journal of Mathematics and Technology, 16(2), 93–106.

[8] Lew, H.-C. (2020). Artificial intelligence (AI) teacher education in Korea. SEAMEO Journal, 2, 119–128. Korea National University of Education.

[9] Lê Thị Hoài Châu & Nguyễn Thế Nhân. (2019). Assessment of 12th graders’ modeling competence in the theme-based teaching “Finding the maximum or minimum value of a function.” Ho Chi Minh City University of Education Journal of Science, 16(12), 891–906. http://journal.hcmue.edu.vn

[10] Mohamed, M. Z., Hidayat, R., Subair, N. N., Sabri, N. B. M., Mahmud, M. K. H. & Baharuddin, S. N. (2022). Artificial intelligence in mathematics education: A systematic literature review. International Electronic Journal of Mathematics Education, 17(3), em0694. https://doi.org/10.29333/iejme/12132

[11] Trần Như Kim & Bùi Anh Kiệt. (2024). Vận dụng mô hình hóa Toán học vào dạy học ứng dụng đạo hàm lớp 12. Journal of Educational Equypment: Applied Research, 2(323), 117–126. https://vjol.info. vn/index.php/tctbgd

[12] Nguyễn Huy Thao, Nguyễn Ngọc Giang, Phạm Huyền Trang, Nguyễn Thị Nga & Dương Minh Tới. (2024). Dạy học ứng dụng “Định lí Sin” vào giải các bài toán thực tiễn nhằm phát triển năng lực giải quyết vấn đề Toán học cho học sinh trung học phổ thông. Vietnam Journal of Education, 24(3), 19–23. https://doi.org/10.31276/vje.2403

[13] Nguyễn Thông Minh & Nguyễn Thị Thúy An. (2023). Tổ chức dạy học môn Toán ở trung học phổ thông theo định hướng phát triển năng lực học sinh. Tạp chí Khoa học Giáo dục Việt Nam, 19(7), 14–21. https:// doi.org/10.15625/2615-8957/12310703

[14] Tashtoush, M. A., Wardat, Y., AlAli, R. & Saleh, S. (2024). Artificial intelligence in education: Mathematics teachers’ perspectives, practices and challenges. Iraqi Journal for Computer Science and Mathematics, 5(1), 60–77. https://doi.org/10.52866/ ijcsm.2024.05.01.004

[15] Trần Vinh. (2020). Lập và sử dụng mô hình với hàm số phù hợp để giải các bài toán thực tế. Vietnam Journal of Education, 492(2), 16–21. https://doi. org/10.31276/vje.492

[16] UNESCO. (2025). Khung năng lực AI cho giáo viên (Phan Văn Quyết, dịch; Nguyễn Văn Giang, hiệu đính). UNESCO Publishing. https://doi.org/10.54675/ MDKU1457

[17] Ứng dụng ChatGPT trong dạy học Toán: Cơ hội và thách thức. (2024). Journal of Educational Equypment: Applied Research, 1(308), 22–33. www. tapchithietbigiaoduc.vn

Articles in Issue