We aim to devise a new approach for model reduction of biochemical reaction networks using their species-reaction graphs, which are bipartite graphs with two sets of nodes, one of which corresponds to the set of species and the other to the different reactions of the network. The set of edges of the graphs includes directed edges from substrate species to the reactions in which they participate, and from the reactions to their constituent product species. The first part of the project consists of devising a new modelling procedure for a reaction network involving the Laplacian matrix and a method for its reduction by considering the Schur complement of the Laplacian obtained by eliminating the rows and columns corresponding to the species and reactions that we wish to eliminate. The second part involves automating the entire procedure in such a way that the user is able to specify the reaction network and the desired extent of reduction and the output is the reduced network.