Continuous Casting is the predominant manufacturing method used in steel production, accounting for 96% of the world’s steel production [1]. In the continuous casting (CC) of steel, tundish is an intermediate vessel that transfers molten steel evenly from the ladle to the mould, with a desired throughput rate and temperature. Acts as a reservoir during ladle change and continues to supply molten steel to the mould. The accurate modelling of tundish metallurgical operations is difficult and computationally expensive [2]. Industries are keenly interested in obtaining reduced-order model (ROMs) for engineering systems, especially for control, optimization, and uncertainty quantification problems. ROMs offer computational efficiency for many query scenarios and are suitable for real-time computations [3]. We are interested in one such application, tundish design in steel industries for the continuous casting process. We are particularly interested in studying steel grade transition during ladle change [5, 6], where multi-phase turbulent flow needs to be simulated for process optimization and the development of efficient control algorithms to find optimum parameters. The computational burden of such flow simulations often calls for developing a reliable reduced-order model for reliable, accurate and fast exploration of parametric space. In this work, we aim to develop data-driven ROM for steel grade intermixing in the tundish during ladle change, and here, we are focusing on investigating both linear subspaces and non-linear subspaces [3, 4] for the development of efficient and accurate ROMs for this application. REFERENCES [1] World Steel Association. World Steel in Figures 2024. [2] D. Mazumdar. Review, analysis, and modeling of continuous casting tundish systems. In steel research international 90.4 (2019), p. 1800279. [3] G. Rozza, G. Stabile, and F. Ballarin. Advanced Reduced Order Methods and Applications in Computational Fluid Dynamics. SIAM Jan 2022. [4] B. Peherstorfer. Breaking the Kolmogorov Barrier with Nonlinear Model Reduction. Notices of the American Mathematical Society, 69(05):1, May 2022. ISSN 1088-9477. [5] M.I.H. Siddiqui, M.H. Kim, Two-Phase Numerical Modeling of Grade Intermixing in a Steelmaking Tundish. Metals 2019, 9, 40. [6] M.I.H. Siddiqui and P.K. Jha, Effect of Inflow Rate Variation on Intermixing in a Steelmaking Tundish during Ladle Change-Over. Steel Research Int. June 2016, 87: 733-744.