The phenomenon of cash-for-vote is a significant threat to the integrity of electoral processes. This paper presents a maximum flow network interdiction model tailored for the Election Commission of India (ECI) to prevent the flow of illicit cash during elections. We formulate the problem mathematically and propose implementation strategies. We model the cash-for-vote scenario as a directed graph \(G=(V,E)\) , where V is the set of nodes representing entities such as political party offices, agents, intermediaries, local distributors, and voters etc., and E is the set of directed edges representing the cash flow routes with its capacities. The political party’s goal is to maximize the flow within the network to influence voters, while the Election Commission of India (ECI) seeks to reduce the maximum cash flow in the distribution network using the least amount of available resources. This work proposes a bilevel optimization problem where one optimization problem serves as a constraint to another. At the upper level, the ECI’s objective is to minimize cash flow, whereas at the lower level, the political party aims to maximize it. The two-stage problem is transformed into a single minimization problem by taking the dual of the inner problem. The resulting problem is then linearized into a mixed integer programming (MIP) problem and solved by standard commercial solvers. The model is tested on the generated realistic data sets that reflects political party’s cash flow at each level of the network. Computational analysis of these test cases offers guidance on the interdiction decisions that the commission should adopt for their efforts to prevent cash flow. We also address the different variants of the model, its mathematical approach to solve and future research directions to pursue.
