Due to the competition of funding among different sectors of the economy and society, Pavement Management Systems (PMS), defined as a systematic tool to manage, plan, assign budget, and programme all the pavement maintenance works, have become necessary for road agencies to desing more cost-effective and longer lasting pavements and to better allocate available funds. Pavement Management System consists of a database that must include the road inventory, the pavement materials, the traffic data and the present

pavement condition data; an analysis module where available information is employed to predict the future condition of the pavements by means of performance models and a reporting module which produces reports to recommend the needed work and how to conduct it. Therefore, performance models are a key factor of any PMS.

The Regional Government of Biscay (RGB), due to its special political status in Spain, have managed since the 80s and is responsible of almost the entire interurban road network in the territory of Biscay, making a total of more than 1.300 km under its control. Aiming to efficiently manage its human and monetary resources, the Regional Government of Biscay has developed its own Pavement Management Systems, with a complete and exhaustive road inventory and a specific plan to introduce pavement structure information in

the database. Moreover, the RGB conducted pavement data collection to know the real state of the entire road network in 2000, 2002 (partially), 2004, 2007, 2011 and 2016, including information about roughness (by means of the International Roughness Index, IRI), skid resistance, deflections and distresses, mainly cracking and rutting.

The aim of this PhD thesis is to develop performance models for IRI and the Side-force coefficient in order to predict the future condition of the road network of Biscay. IRI models are developed for new two-lane roads with flexible and semi-rigid pavements. For flexible pavements, influencing factors are the age of the pavement, the total thickness of bituminous layers and the accumulated number of heavy vehicles that circulated on the project lane. For semi-rigid pavements, the model requires the age of the pavement, the

accumulated number of both heavy and total vehicles that passed over the pavement and a factor for considering the non bituminous materials. The proposed model for forecasting skid resistance can be deployed in any type of road, two-lane roads or double carriageway roads, and for all the usual materials employed in surface layers in Biscay: asphalt concrete, discontinuous mixes, porous asphalts and slurries. The model has a short version which includes as variables the Annual Average Daily Traffic (AADT) of the year of the data collection, a factor which considers the type of surface material and a factor for considering

the type of road and the number of lanes per direction. Apart from these variables, the long version of the model also includes a factor that combines the AADT, the type of road and number of lanes, the surface material and the required Polished Stone Value according to Spanish regulations. The introduction of the Annual Average Daily Traffic of all the traffic volume and not the AADT of the heavy traffic represents a difference with the previously proposed models.