Control estadístico y modelos de regresión linealuna forma práctica de control de puentes

Resum

This Thesis provides a process for transform data obtained by Monitoring Bridges into information. This information will be useful not only in terms of conservation and maintenance, but also for the designer that when accessing the database records can feed back their calculation models leading to what is already known as meta-model calculations or even review existing regulations (much to the side of safety in cases where no experimental data). This procedure assumes validate data and create models from the Monitoring. Later, during field inspections, these models have to be updated gradually. The new field data will both check the progress of the measure (Statistical Control) and adjust the initial model (Prediction Bands), thus defining new scenarios (infinite, a priori) the state of the bridge. Finally, the effectiveness of different management tactics at the network level bridges (Markov Chains) can simulate. The statistics tools presented are not new but what is new is the global and inclusive vision for maximum use of the experience of monitoring in the field of engineering for the new century.