Automated Synthesis Of Product Return Recommendations Via Groq And Large Language Models
DOI:
https://doi.org/10.24269/mtkind.v20i1.13659Keywords:
after-sales service automation, deterministic reasoning architecture, Large Language Models (LLM), Language Processing Unit (LPU), Risk Priority Number (RPN)Abstract
Industrial economic resilience depends on the efficiency of after-sales service provisioning, which is often hindered by semantic ambiguity in customer reports and latency constraints of conventional computing infrastructures. This study examines the integration of a Language Processing Unit (LPU) with a Large Language Model (LLM) under a Deterministic Reasoning Architecture (DRA) framework to address these limitations. Experiments were conducted on a heterogeneous dataset (N = 27.500) consisting of operational service records from PT Rekaindo Global Jasa and a Southeast Asian manufacturing entity over the period 2021–2025. Semantic complexity analysis based on Shannon Entropy indicates that the Repair category exhibits the highest information density (5.2 bits), corresponding to an increased risk of logical failure. Performance benchmarking demonstrates that the proposed LPU-based architecture achieves deterministic inference with a Risk Priority Number (RPN) of 42 significantly lower than stochastic GPU-based baselines (RPN > 120). Predictive integrity evaluation yields an AUC–ROC of 0.988 and an inter-rater agreement of 0.81 (Fleiss Kappa), indicating substantial alignment between automated recommendations and expert assessments. Economic robustness is validated through Monte Carlo simulations, showing a 94.2 % probability of achieving Return on Investment within 20 months, even under high-volatility scenarios. Furthermore, the framework complies with ISO/IEC 42001:2023 and the EU AI Act, achieving a Fairness Ratio above 0.94. Overall, the results demonstrate that the LPU–LLM synergy enables fast, reliable, and responsible generative AI deployment in industrial settings.
Downloads
References
[1] C. Gao, X. Hu, S. Gao, X. Xia, and Z. Jin, “The Current Challenges of Software Engineering in the Era of Large Language Models,” ACM Transactions on Software Engineering and Methodology, vol. 34, no. 5, pp. 1–30, 2025, doi: 10.1145/3712005.
[2] Y. Chang et al., “A Survey on Evaluation of Large Language Models,” ACM Transactions on Intelligent Systems and Technology, vol. 15, no. 3, pp. 1–45, 2024, doi: 10.1145/3641289.
[3] G. Huang, Y. Long, Y. Li, and G. Papanastasiou, “From explainable to interpretable deep learning for natural language processing in healthcare: How far from reality?,” Computational and Structural Biotechnology Journal, 2024, doi: 10.1016/j.csbj.2024.05.004.
[4] L. Song et al., “The Early Bird Catches the Leak: Unveiling Timing Side Channels in LLM Serving Systems,” IEEE Transactions on Information Forensics and Security, vol. 20, pp. 11431–11446, 2024, doi: 10.1109/tifs.2025.3622954.
[5] H. Zhao et al., “Explainability for Large Language Models: A Survey,” ACM Transactions on Intelligent Systems and Technology, vol. 15, no. 2, pp. 1–38, 2024, doi: 10.1145/3639372.
[6] Y. Yu et al., “Fine-Tuning Large Language Models to Improve Accuracy and Comprehensibility of Automated Code Review,” ACM Transactions on Software Engineering and Methodology, vol. 34, no. 1, pp. 1–26, 2024, doi: 10.1145/3695993.
[7] K. Lu et al., “Q-Infer: Towards Efficient GPU-CPU Collaborative LLM Inference via Sparsity-Aware Dynamic Scheduling,” ACM Transactions on Architecture and Code Optimization, 2025, doi: 10.1145/3764589.
[8] M. Landers, “Adapting to the Unsanctioned Use of AI-Supported Technologies in Student Assessments,” Higher Education for the Future, 2024, doi: 10.1177/23476311241300608.
[9] A. George, E. J. Muttom, J. Joyan, N. Binoy, J. Shogi, and R. Thomas, “A Personalized AI Assistant for Analyzing and Authoring Product Reviews using Groq/Llama3-70B,” in 2025 International Conference on Computing Technologies & Data Communication (ICCTDC), 2025, pp. 1–6. doi: 10.1109/ICCTDC64446.2025.11158767.
[10] S. Lambiase, G. Catolino, F. Palomba, F. Ferrucci, and D. Russo, “Investigating the Role of Cultural Values in Adopting Large Language Models for Software Engineering,” ACM Transactions on Software Engineering and Methodology, vol. 1, no. 1, 2024, doi: 10.1145/3678172.
[11] C. Nyongesa, M. Hogarth, and J. Pa, “Artificial intelligence-driven natural language processing for identifying linguistic patterns in Alzheimer’s disease and mild cognitive impairment: A study of lexical, syntactic, and cohesive features of speech through picture description tasks,” Journal of Alzheimer’s Disease, vol. 106, pp. 120–138, 2025, doi: 10.1177/13872877251339756.
[12] Y. Bestgen, “Estimating lexical diversity using the moving average type-token ratio (MATTR): Pros and cons,” Research Methods in Applied Linguistics, 2025, doi: 10.1016/j.rmal.2024.100168.
[13] D. Friedman and A. Dieng, “The Vendi Score: A Diversity Evaluation Metric for Machine Learning,” Trans. Mach. Learn. Res., vol. 2023, 2022, doi: 10.48550/arxiv.2210.02410.
[14] B. Cosenza, L. Crisci, G. Amati, and M. Turisini, “miniLB: Benchmarking Lattice Boltzmann simulations on AMD, Intel, and NVIDIA GPUs,” Future Gener. Comput. Syst., vol. 175, p. 108032, 2025, doi: 10.1016/j.future.2025.108032.
[15] E. Husom et al., “Sustainable LLM Inference for Edge AI: Evaluating Quantized LLMs for Energy Efficiency, Output Accuracy, and Inference Latency,” ACM Transactions on Internet of Things, 2025, doi: 10.1145/3767742.
[16] X. Wang, X. Hui, C. Liao, and X. Shen, “Reductive Analysis with Compiler-Guided Large Language Models for Input-Centric Code Optimizations,” Proc. ACM Program. Lang., vol. 9, no. PLDI, Jun. 2025, doi: 10.1145/3729282.
[17] M. Hussain, N. Mirjat, F. Shaikh, L. L. Dhirani, S. A. Khatri, and L. Kumar, “Reliability Assessment of Wind Turbines Based on Failure Rate and Downtime Estimation,” International Journal of Energy Research, 2025, doi: 10.1155/er/3424586.
[18] L. Bojic et al., “Comparing large Language models and human annotators in latent content analysis of sentiment, political leaning, emotional intensity and sarcasm,” Scientific Reports, vol. 15, 2025, doi: 10.1038/s41598-025-96508-3.
[19] R. Miller, H. Whelan, M. Chrubasik, D. Whittaker, P. Duncan, and J. Gregório, “A Framework for Current and New Data Quality Dimensions: An Overview,” Data, vol. 9, no. 12, p. 151, 2024, doi: 10.3390/data9120151.
[20] B. Jin, G. Liu, C. Han, M. Jiang, H. Ji, and J. Han, “Large Language Models on Graphs: A Comprehensive Survey,” IEEE Transactions on Knowledge and Data Engineering, vol. 36, no. 12, pp. 8622–8642, 2024, doi: 10.1109/tkde.2024.3469578.
[21] A. Chokami and G. Rabitti, “An Exact Game-Theoretic Variable Importance Index for Generalized Additive Models,” Journal of Computational and Graphical Statistics, vol. 33, pp. 1276–1285, 2024, doi: 10.1080/10618600.2024.2327577.
[22] H. Tan, Q. Luo, J. Li, and Y. Zhang, “LLM4Decompile : Decompiling Binary Code with Large Language Models,” 2023.
[23] R. Diaz Milian, P. Moreno Franco, W. D. Freeman, and J. D. Halamka, “Revolution or Peril? The Controversial Role of Large Language Models in Medical Manuscript Writing,” Mayo Clinic Proceedings, vol. 98, no. 10, pp. 1444–1448, 2023, doi: 10.1016/j.mayocp.2023.07.009.
[24] G. Biagini, “Towards an AI-Literate Future: A Systematic Literature Review Exploring Education, Ethics, and Applications,” International Journal of Artificial Intelligence in Education, 2025, doi: 10.1007/s40593-025-00466-w.
[25] V. Vimbi, N. Shaffi, and M. Mahmud, “Interpreting artificial intelligence models: a systematic review on the application of LIME and SHAP in Alzheimer’s disease detection,” Brain Informatics, 2024, doi: 10.1186/s40708-024-00222-1.
[26] Z. Wang et al., “M2CVD: Enhancing Vulnerability Understanding through Multi-Model Collaboration for Code Vulnerability Detection,” ACM Trans. Softw. Eng. Methodol., Oct. 2025, doi: 10.1145/3771923.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 usman nurhasan, Dian Hanifudin Subhi, Ibnu Tsalis Assalam
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
MULTITEK is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License
