Deciding on maintenance of small municipal roads based on GIS simplified procedures

Abstract

This paper describes a survey of a Portuguese municipality road network and the development of a simplified procedure to support road network maintenance. This procedure considers inspection and quality evaluation of pavements. In addition, it allows programming of maintenance actions based on deterministic and probabilistic models to predict the evolution of pavement condition. The potential of this procedure is demonstrated through its use to establish scenarios, which enable the development of maintenance programs for the municipality’s road network.The data was represented in a commercial geographic information system (GIS) and processed through Python scripts. The GIS capabilities can analyse, query and modify the road network database as well as the production of maps and useful graphical information.The methodology was applied to 55.7 km of routes from a small municipality in Portugal, aiming at demonstrating the practical feasibility of the procedure, even when specialized means are not available. This means that after defining the rules for the collection and recording of information, it is not essential to maintain a specialized technical structure to proceed with the process of network analysis and decision-making.Two different techniques provided the prediction of pavement condition. For segments in which traffic of heavy vehicles was significant, the deterministic AASHTO model was applied. For the remaining cases, Markov chains (a probabilistic model) were used.The study made it possible to carry out some analysis on future scenarios of maintenance on the network, providing better decisions for the overall municipality and/or for specific areas of its territory. The obtained scenarios resulted in a prediction of pavement condition over the planning period, as well as a cost estimate for the studied scenarios.