The main purpose of the paper is to develop a tool to effectively support the transport infrastructure planning process. The novel method is based on modified ant colony optimization and is used for evaluating new infrastructure sections in the public transport. The section could contain a single route as well as a combination of two or more routes. The authors propose a two-step evaluation. Step one uses a modified ant colony optimization to evaluate a single underground route. As a result, for each route, the effort that artificial ants have to cover the routes is obtained. Such effort is calculated considering the distance, total length, number of inhabitants, and areas with access to the rail network. In step two, the proposed routes are combined to create variants. The evaluation of combining routes use additional parameters like crossing time or travel time in alternative means. The case study located in one of the biggest agglomerations in Poland shows the method's utility in the evaluation of options for a railway tunnel. The analyses show that all criteria have an influence on the results and that the new, two steps method gives an intriguing effect. The presented methodology contains novel elements and their implementation to specific transport infrastructure elements. The results are contrasted with an algorithm based on multi-criteria analysis, which showed the greater complexity of the proposed approach.