{ "article_id": 1648, "public_id": "ca899497-1652-4d61-98cc-f7e4095feba0", "internal_identifier": null, "source_identifier": null, "status": "published", "type": "research_article", "title": "Business-Aware Agentic Forecast Refinement for Enterprise Financial Planning", "title_secondary": null, "abstract": "Enterprise financial forecasts are generated across multiple business dimensions such as products, regions, channels, customer cohorts, and segments. In a large enterprise, this can result in billions of populated intersections that finance teams cannot review manually before a planning deadline. This paper presents a proactive, human-supervised agentic AI workflow implemented in a financial-technology and payments planning environment. The process starts when a baseline forecast reaches a configured planning milestone. The agent reviews each granular trajectory, identifies relevant historical periods, considers business seasonality, selects an adjustment strategy, generates the correction, and validates the resulting forecast before proposing the adjustment to finance users. The implementation was evaluated in one quarterly forecast and one long-range planning cycle covering 11 dimensions, approximately 1.2 billion populated intersections, and more than 15 financial planners. Planners accepted 67.2% of the proposed adjustments. For the quarterly forecast, the organization’s forecast-to-actual accuracy measure improved from approximately 96% to 98%. This workflow shifted planner effort from searching for anomalies to reviewing proposed corrections, applying business judgment, and collaborating on forward-looking assumptions while preserving finance authority over final decisions.", "language": "en", "english_title": null, "english_abstract": null, "doi": "10.36287/setsci.25.7.001", "doi_url": "https://doi.org/10.36287/setsci.25.7.001", "resource_url": "https://www.set-science.com/a/ca899497-1652-4d61-98cc-f7e4095feba0", "pdf_url": "https://www.set-science.com/a/ca899497-1652-4d61-98cc-f7e4095feba0/pdf", "og_image": "https://www.set-science.com/assets/img/og-card-v2.png", "authors": [ { "public_id": "c772f625-3300-4c76-bcdc-91132e65f6b5", "given": "Hari", "family": "Rajagopalan", "full": "Hari Rajagopalan", "literal": null, "orcid": null, "orcid_url": null, "email": "hari.buvanesh@gmail.com", "corresponding": true, "sequence": "first", "affiliations": [ { "name": "Systems Architect, Stripe, LLC., Austin, TX, USA", "department": null, "country": null, "country_code": null, "ror": null, "ror_url": null } ] } ], "keywords": [ "Agentic AI", "enterprise forecasting", "FP&A", "human-in-the-loop systems", "anomaly detection", "forecast adjustment", "decision intelligence", "financial technology" ], "translations": [], "subjects": [ { "name": "Finance", "label": "Finance", "path": "finance", "is_primary": true } ], "subject_names": [ "Finance" ], "published_date": "2026-10-06", "published_year": "2026", "received_date": null, "accepted_date": null, "volume": "25", "issue": null, "page_first": "37", "page_last": "41", "article_number": "7", "conference": { "short_name": "ICOF2026", "full_title": "International Conference on Open Finance", "name": "International Conference on Open Finance", "acronym": "ICOF2026", "location": "Boston, United States", "start_date": "2026-09-24", "end_date": "2026-09-24", "public_id": "0181a00a-2de1-4dcf-9458-2c9fb1cfd6ed" }, "proceedings": { "title": "Proceedings of the International Conference on Open Finance (ICOF 2026)", "series_title": "SETSCI Conference Proceedings", "publisher": "SET Technology", "isbn": "978-625-95366-7-5", "eisbn": "978-625-95366-7-5", "issn": null, "eissn": "2687-5527", "doi": "10.36287/setsci.2026.25", "doi_prefix": "10.36287", "slug": "proceedings-of-the-international-conference-on-open-finance-icof-2026", "public_id": "99f5c14a-5332-4630-9f4e-19033533a996", "url": "https://www.set-science.com/p/proceedings-of-the-international-conference-on-open-finance-icof-2026" }, "license": { "name": "CC BY 4.0", "url": "https://creativecommons.org/licenses/by/4.0/", "code": "CC-BY-4.0", "open_access": true }, "copyright": { "holder": "SET Technology", "year": "2026" }, "funding_statement": null, "funders": [], "references": [ { "order": 1, "raw": "R. J. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice, 3rd ed. Melbourne, Australia: OTexts, 2021. [Online]. Available: https://otexts.com/fpp3/", "doi": null, "title": null, "authors": null, "source": null, "year": 2021, "volume": null, "pages": null }, { "order": 2, "raw": "S. L. Wickramasuriya, G. Athanasopoulos, and R. J. Hyndman, “Optimal forecast reconciliation for hierarchical and grouped time series through trace minimization,” Journal of the American Statistical Association, vol. 114, no. 526, pp. 804–819, 2019.", "doi": null, "title": null, "authors": null, "source": null, "year": 2019, "volume": null, "pages": null }, { "order": 3, "raw": "V. Chandola, A. Banerjee, and V. Kumar, “Anomaly detection: A survey,” ACM Computing Surveys, vol. 41, no. 3, article 15, 2009, doi:10.1145/1541880.1541882.", "doi": "10.1145/1541880.1541882", "title": null, "authors": null, "source": null, "year": 2009, "volume": null, "pages": null }, { "order": 4, "raw": "I. Okpala, A. Golgoon, and A. R. Kannan, “Agentic AI systems applied to tasks in financial services: Modeling and model risk management crews,” arXiv:2502.05439, 2025.", "doi": null, "title": null, "authors": null, "source": null, "year": 2025, "volume": null, "pages": null }, { "order": 5, "raw": "M. Dimodugno and M. Mammadov, “AICrypto-Assistant: A multi-agent LLM platform for democratizing crypto-asset analysis,” SETSCI Conference Proceedings, vol. 24, pp. 17–22, 2025, doi:10.36287/setsci.24.2.017.", "doi": "10.36287/setsci.24.2.017", "title": null, "authors": null, "source": null, "year": 2025, "volume": null, "pages": null }, { "order": 6, "raw": "S. Amershi et al., “Guidelines for human-AI interaction,” in Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, paper 3, pp. 1–13, 2019, doi:10.1145/3290605.3300233.", "doi": "10.1145/3290605.3300233", "title": null, "authors": null, "source": null, "year": 2019, "volume": null, "pages": null }, { "order": 7, "raw": "R. Sethi et al., “Presto: SQL on everything,” in 2019 IEEE 35th International Conference on Data Engineering, pp. 1802–1813, 2019, doi:10.1109/ICDE.2019.00196.", "doi": "10.1109/ICDE.2019.00196", "title": null, "authors": null, "source": null, "year": 2019, "volume": null, "pages": null } ], "statements": { "acknowledgements": null, "conflict_of_interest": null, "data_availability": null, "ai_disclosure": null }, "credit": [], "history": { "published": "2026-10-06" } }