Evaluating the filtration performance of a filter design traditionally relies on a global breakthrough curve, which represents changes in outlet concentration over time. This curve does not account for adsorption phenomena inside the filter, often necessitating trial-and-error experimentation for optimizing filter shape. To address this issue, we introduce an innovative approach that employs computational fluid dynamics (CFD) to define local breakthrough curves. These curves represent concentration changes over time at specific points within the filter. By comparing these local breakthrough curves with the conventional global breakthrough curve, we aim to enhance the precision of filter designs. This optimization strategy seeks to achieve shapes that meet predetermined performance targets or improve the efficiency of existing filters. Utilizing CFD-generated local breakthrough curves in the optimization process enables the prediction of essential filter elements required to meet specified performance criteria, such as height and thickness. Moreover, this method identifies underperforming areas in existing filters and suggests strategic avenues for improvement. Our results underscore the utility of CFD-informed local breakthrough curves as invaluable tools for streamlining the filter design optimization process.