Development of artificial neural network for condition assessment of bridges based on hybrid decision making method – Feasibility study

•An original ANN application used as a bridge management system is presented.•Neural network was used to rank bridges on the basis of their technical condition.•Performance of the model is shown on the example of railway bridges.•Comprehensive tests are verified with the use of novel hybrid multi criteria method. Managing a bridge at an appropriate level of reliability requires knowledge of its technical condition, which is decisive in terms of maintenance and repair activities. This is a multi-criteria decision-making problem which results from the need to allocate limited financial resources to this work. Although many calculation models have been suggested in published sources, none of them has ever met these requirements. The algorithm presented by the authors allows for the assessment of any number of bridges, taking into account the diversity of solutions in terms of materials and structures, and can provide a solution to this problem. This hybrid calculation model, combining the modified Extent Analysis Fuzzy Analytic Hierarchy Process (EA FAHP) and Dominant Analytic Hierarchy Process (DAHP), has been verified on many existing bridges. However, due to the advanced mathematical algorithm, its practical use may create some difficulty for engineers applying it in practice. To avoid implementation problems, the authors have proposed a neural network model to facilitate the work of the staff operating the bridges. The application of the proposed model was presented in an article about a group of selected railway bridges that have been operating in Poland for over 100 years. The objects selected for analysis have open decks, which means that they are more susceptible to damage compared to road bridges. The accuracy of the artificial neural network (ANN) model was verified by comparing the results provided with the results obtained using the aggregated EA FAHP + DAHP model.