Due to the population densities in large cities, efficient urban transport systems are essential in solving traffic problems. Unfortunately, it cannot be based on individual transport, because this quickly leads to city congestion, air pollution etc. A good solution is an efficient and optimized public transport. One of the ideas that can realize this aim is the traffic organization in the form of a hub&spoke system. In this paper, we present an idea, based on hypergraphs and evolutionary optimization to apply elements of the hub&spoke transportation system using a hypergraph transversal structure. The evolutionary algorithm is used to obtain this structure. It allows for an easy use of various quality criteria and additional information that data structures describing such graphs can contain, e.g. weights of edges with information on bandwidth, connection time, number of travelers, etc. We present the computed results based on the publicly available timetable of the Warsaw Public Transport system.