This paper defines a capacitated multiple-source multiple-sink shortest path problem and introduces its extension, called the capacitated multiple-source multiple-sink shortest path network interdiction problem (CMSSNIP). CMSSNIP examines the actions of attackers who attempt to maximize the total shortest path of network users trying to reach the crime locations for the aid process after causing an incident in certain regions to provide strategic information for the defense systems of the government. In this context, the exact mathematical model ensuring useful information about safe routes to network users is proposed. In this manner, to the best knowledge of authors, the CMSSNIP consisting of multiple-source nodes and multiple-sink nodes and considering capacity-demand relations between security units and crime locations is studied for the first time. Accordingly, a set of scenarios is considered based on the interdiction budget levels and the number of crime locations through a real case application to show the applicability of the model. Furthermore, computational experiments are performed to evaluate the model performance on different-sized networks. It is realized that especially the model provides resilient strategies versus interdictions in terms of obtaining the safe shortest paths at the operational level within seconds in the real case application.