Conformal Prediction for Enterprise Database Capacity Planning

The central idea of this paper is simple: database planners need more than one expected capacity date. Using SARIMAX with selected business drivers and residual-based split conformal prediction, the study showed how a 6.2-unit threshold can be converted into a practical window from March 16, 2026, to May 3, 2026, with the central forecast falling on April 9, 2026. This range is the main outcome of the work. It gives infrastructure and business teams a better way to discuss procurement timing, budget approval, and capacity risk because it accepts that database growth cannot be predicted as one exact date. As discussed in Section 7, the next step is being explored as a small MVP or proof of concept, where planners can use a chatbot-style interface to change business assumptions and see how the conformal capacity window moves. This can form the basis for a continuing paper on agentic AI-supported capacity planning.

[1]     R. J. Hyndman and G. Athanasopoulos, Forecasting: Principles and Practice, 3rd ed. Melbourne, Australia: OTexts, 2021. [Online]. Available: https://otexts.com/fpp3/. Accessed: Jun. 11, 2026. [2]     V. Vovk, A. Gammerman, and G. Shafer, Algorithmic Learning in a Random World. New York, NY, USA: Springer, 2005. [3]     A. N. Angelopoulos and S. Bates, “A gentle introduction to conformal prediction and distribution-free uncertainty quantification,” arXiv:2107.07511, 2021. [Online]. Available: https://arxiv.org/abs/2107.07511. Accessed: Jun. 11, 2026. [4]     E. Tabassi, Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1. Gaithersburg, MD, USA: National Institute of Standards and Technology, Jan. 2023. [Online]. Available: https://doi.org/10.6028/NIST.AI.100-1. Accessed: Jun. 11, 2026. [5]     IBM Corporation, “Capacity management and capacity planning concepts,” IBM Documentation. [Online]. Available: https://www.ibm.com/docs/. Accessed: Jun. 11, 2026. [6]     Amazon Web Services, AWS Well-Architected Framework. Seattle, WA, USA: Amazon Web Services, Inc. [Online]. Available: https://docs.aws.amazon.com/wellarchitected/. Accessed: Jun. 11, 2026. [7]     B. Lim, S. Ö. Arık, N. Loeff, and T. Pfister, “Temporal Fusion Transformers for interpretable multi-horizon time series forecasting,” International Journal of Forecasting, vol. 37, no. 4, pp. 1748–1764, Oct.–Dec. 2021. [8]     J. Gawlikowski et al., “A survey of uncertainty in deep neural networks,” Artificial Intelligence Review, vol. 56, pp. 1513–1589, 2023. [9]     European Parliament and Council of the European Union, Regulation on Artificial Intelligence: European Union Artificial Intelligence Act. Brussels, Belgium: European Union, 2024. [Online]. Available: https://artificialintelligenceact.eu/. Accessed: Jun. 11, 2026. [10] Organisation for Economic Co-operation and Development, OECD AI Principles. Paris, France: OECD. [Online]. Available: https://oecd.ai/en/ai-principles. Accessed: Jun. 11, 2026.