Harnessing Generative AI for Financial Risk Anticipation in Retail and Coatings Manufacturing Supply Chains
DOI:
https://doi.org/10.5281/zenodo.20444175Keywords:
Financial Risk, Generative AI, Predictive Modelling, Supply Chain Finance, Supply Chain Management, Retail Business, Paint Industry, Chain Vertical Integration, Time of Day Effect.Abstract
Financial risk is an increasingly complex challenge in supply chains. Dynamic changes in the market affect multiple stakeholders across different tiers of the supply chain. Accurately defining and estimating these risks is crucial for optimizing working capital utilization and planning. Generative AI has been identified as a technology that can play a significant role in operational processes. Research on its application has focused primarily on operational processes, while the potential benefits for predictive modelling, particularly in supply chain financial risk management, remain unexplored. This study attempts to address this gap by proposing a framework for simulating various financial risk exposures in a retail supply chain and cross-tide testing the resulting specials within the supply chain of a paint manufacturer.
The proposed framework comprises three main components: (a) establishing a data architecture with the required data elements, sources, and flows for preparing the data, (b) implementing generative AI models to simulate financial risk exposures, and (c) defining metrics for validating the output of the simulated models. These components are examined in detail, and their implementation is illustrated through a retail use case and a cross-supply chain application involving a paint manufacturer. In the retail case, financial risks related to stock levels, turnover, working capital, and liquidity stress under demand volatility are explored. The second case involves a paint manufacturer faced with a potential price increase of key raw material. It investigates the impact of sourcing decisions underprice uncertainty on raw material costs and financial exposure.
References
[1] Alghanim, A. (2023). Predictive analytics and machine learning for real-time supply chain risk mitigation and agility. Sustainability, 15(20), 15088. doi:10.3390/su152015088
[2] Uniocular, E., Kabadurmuş, O., & Yıldızbaşı, A. (2024). Evaluation of risk strategies for supply chain sustainability: A comprehensive review and future directions. Helion, 10, e****. Doi: 10.1016/j.heliyon.2024. ****
[3] Bahroun, Z., Kabadurmuş, O., & Yıldızbaşı, A. (2025). A systematic analysis of generative artificial intelligence applications in supply chain management. Cleaner Logistics and Supply Chain, 9, 100088. doi:10.1016/j.clscn.2025.100088
[4] Bani-Irshid, A. H., & (co-authors). (2024). Supply chain risk assessment and mitigation under the COVID-19 pandemic: A review. International Journal of Production Management and Engineering, 12(1), –.
[5] Brintrup, A., (co-authors). (2024). Digital supply chain surveillance using artificial intelligence. International Journal of Production Research, 62(**), –. doi:10.1080/00207543.2023.2270719
[6] Chang, V., (co-authors). (2024). Credit risk prediction using machine learning and deep learning approaches. Risks, 12(11), 174. doi:10.3390/risks12110174
[7] Chatterjee, S., (co-authors). (2025). GAN-based synthetic time-series data generation for improved forecasting and anomaly detection (CWGAN-GP framework). Expert Systems with Applications, 2*, * –. Doi: 10.1016/j.eswa.2024. ****
[8] Chen, L., (co-authors). (2025). Generative adversarial synthetic neighbors-based unsupervised anomaly detection (GASN). Scientific Reports, 15, 84863. doi:10.1038/s41598-024-84863-6
[9] Collier, Z. A., (co-authors). (2025). How good are large language models at product risk assessment? Risk Analysis, 45(**), –. doi:10.1111/risa.14351
[10] Dewi, A. C., (co-authors). (2024). A novel model for sustainable supplier selection in the paints and coatings industry. Sustainability, 16(19), 8718. doi:10.3390/su16198718
[11] Dogariu, M., & Ștefan, L. (2022). Generation of realistic synthetic financial time-series. ACM Transactions on Multimedia Computing, Communications, and Applications, 18(**), 1–27. doi:10.1145/3501305
[12] Fosso Wamba, S., (co-authors). (2024). The interplay between artificial intelligence, production, and operations management: Implications for risk and resilience. International Journal of Production Research, 62(**), –. doi:10.1080/00207543.2024.2321826
[13] Gates, N., (co-authors). (2025). From traditional scoring to explainable machine learning in credit risk assessment: A review. SSRN Working Paper.
[14] Gregnanin, M., (co-authors). (2025). A generative adversarial graph neural network for synthetic financial time series (Sig-Graph GAN). In Proceedings of the ** Workshop/Conference (pp. –).
[15] Han, C., (co-authors). (2025). Time-series-based anomaly detection in industrial control systems with semi-supervised GAN data augmentation. Processes, 13(9), 2885. doi:10.3390/pr13092885
[16] Jackson, I., Ivanov, D., Dolgui, A., & Namdar, J. (2024). Generative artificial intelligence in supply chain and operations management: A capability-based framework for analysis and implementation. International Journal of Production Research, 62(**), 1–26. doi:10.1080/00207543.2024.2309309
[17] Joglekar, N., (co-authors). (2025). An AI-driven approach to assess sentiments and interpret supply chain risk signals. International Journal of Production Research, 63(**), –. doi:10.1080/00207543.2025.2532142
[18] Kong, Y., (co-authors). (2024). Large language models for financial and investment management: Opportunities, limitations, and risk considerations. The Journal of Portfolio Management, 51(2), 211–**.
[19] Liang, Y., (co-authors). (2023). Derisk: A deep learning risk prediction framework for real-world credit risk prediction. arXiv preprint arXiv:2308.03704.
[20] Lin, K.-Y., Wu, S.-Y., (co-authors). (2025). Generative AI-driven resilience in supply chain management: UNISONE framework for disruption modelling and capacity optimization. International Journal of Production Research, 63(**), –. doi:10.1080/00207543.2025.2574412
[21] Lim, W., (co-authors). (2024). Future of generative adversarial networks (GANs) for anomaly detection: A systematic review and research agenda. Decision Support Systems, 1*,* –. doi:10.1016/j.dss.2024.****
[22] Liu, Y., (co-authors). (2025). AT-FinGPT: Financial risk prediction via an audio-text large language model. Journal of Financial Data Science, (), –. doi:10.1016/j.jfds.2025.**
[23] Long, Y., (co-authors). (2025). Leveraging synthetic data to tackle machine learning challenges in operations and supply chains. International Journal of Production Research, 63(**), –. doi:10.1080/00207543.2024.2447927
[24] Montevechi, A. A., (co-authors). (2024). Advancing credit risk modelling with machine learning: A comprehensive review. Engineering Applications of Artificial Intelligence, 1*,* 105***. doi:10.1016/j.engappai.2024. ****
[25] Qian, C., (co-authors). (2025). RGAnomaly: Data reconstruction-based generative anomaly detection for time series in IoT. Future Generation Computer Systems, 1*,* –. doi:10.1016/j.future.2025.****
[26] Rinaldi, M., (co-authors). (2022). A literature review on quantitative models for supply chain risk management. Annals of Operations Research, 315(1), –. doi:10.1007/s10479-022-04700-9
[27] Shen, F., (co-authors). (2021). A new deep learning ensemble credit risk evaluation model for imbalanced data. Applied Soft Computing, 1*,* 107***. doi:10.1016/j.asoc.2020.107***
[28] Sikharulidze, M., & Fuschi, D. L. (2023). Causes of paints and coatings raw materials supply chain crisis and its impact on businesses. *Journal/Proceedings Title, *(), –.
[29] Suhadolnik, N., (co-authors). (2023). Machine learning for enhanced credit risk assessment. Journal of Risk and Financial Management, 16(12), 496. doi:10.3390/jrfm16120496
[30] Takahashi, T., & Mizuno, T. (2025). Generation of synthetic financial time series by diffusion models. Quantitative Finance, 25(10), 1–10. doi:10.1080/14697688.2025.2528697
[31] Vidal, U., Obregon, M., Ramos, E., Verma, R., & Coles, P. S. (2024). Sustainable and risk-resilient circular supply chain: A Peruvian paint manufacturing supply chain model. Sustainable Futures, 7, 100207. doi:10.1016/j.sftr.2024.100207
[32] Wyrembek, M., (co-authors). (2025). Causal machine learning for supply chain risk prediction. International Journal of Production Research, 63(**), –. doi:10.1080/00207543.2025.2458121
[33] Xie, Q., (co-authors). (2024). Financial challenges in large language models (FinLLMs): Benchmarking, evaluation, and risk of leakage. In Proceedings of the FinNLP Workshop at ACL 2024 (pp. –).
[34] Yang, M., Lim, M. K., Qu, Y., Ni, D., & Xiao, Z. (2023). Supply chain risk management with machine learning technology: A literature review and future research directions. Computers & Industrial Engineering, 175, 108859. doi:10.1016/j.cie.2022.108859
[35] Zheng, G., Kong, L., & Brintrup, A. (2023). Federated machine learning for privacy preserving, collective supply chain risk prediction.
Additional Files
Published
Data Availability Statement
None
