Asymmetric multilayered filters, comprising a series of membranes with varying pore sizes stacked on top of one another, allow filtration to be tailored in a variety of novel ways. We develop a network model that systematically captures the complex filtration behaviour in such multilayer filters. The model allows us to understand the response of the system when challenged with a particular feed composition, characterized through the particle size and adhesivity to the membrane. We show how the model enables comprehensive and time-efficient sweeps in parameter space to be conducted that determine the optimal multilayered filter configuration for a given filtration challenge, classified by the number of membrane layers, the change in pore size between each layer (filter taper angle), and the level of trans pore interconnectivity between each layer. The model allows us to isolate and analyse the effect of each of the specific filter characteristics and identify the practical merits and disadvantages. In particular, we predict that the optimal arrangement for maximizing throughput through the filter is to have pore radius gradually decreasing with depth, and a slight level of pore interconnectivity, with the precise set-up a function of the particle size, adhesivity and number of filter layers. The results of the analysis are used to draw conclusions on the design of membrane filters for optimal filter performance. (C) 2016 Published by Elsevier B.V.