A novel operational filtration index to support decision-making in wine bottling

The final filtration stage in wine bottling is critical for ensuring microbiological quality but is often affected by high variability and costs due to filter fouling. Traditional filterability indices lack physical interpretability and fail to account for key operational factors such as batch size and filtration area. This study introduces the Operational Filtration Index (OFI), derived from a hyperbolic model that describes the filtered volume over time under constant pressure, assuming standard membrane blocking. The Operational Filtration Index (OFI) quantifies the volume filtered when the flow rate drops to 25% of its initial value, which corresponds to a 75% obstruction of the effective filter surface under the assumption of direct proportionality between flow rate and active filtration area. Based on 46 controlled experiments with volume recorded every second, the model was fitted using nonlinear optimization. Integer optimization identified five measurement points that achieved a maximum error below 5%. External validation with 14 independent curves confirmed the model's consistency (R2 = 0.999) and low error levels. The OFI outperformed classical indicators in both discriminative power and stability. When adjusted by filtration area and combined with slack calculation, it supports operational decision-making by improving batch allocation and filtration line management. By providing a physically meaningful, replicable framework, the area-adjusted OFI helps close the gap between theoretical model development and practical implementation. This integrated approach bridges the gap between theoretical modeling and industrial practice, offering a reliable and physically interpretable index to guide filtration strategies in wine bottling operations.