Effect of Spatial Variability on Maintenance and Repair Decisions for Concrete Structures

Due to the increasingly number of elder and deteriorating structures, maintenance is becoming a serious and more complex problem in most of the countries. A lot of studies have been carried out in this area for years. However, the fact that a lot of parameters show spatial random variability, which is linked with dependencies on temperature, w/c ratio, cement type, humidity and workmanship, is not included explicitly in most of the previous studies. Therefore, the main objective of the project is to develop a maintenance cost prediction model that helps to determine the optimal maintenance strategy with special consideration of the spatial variability of concrete degradation. The scope of the research is limited to concrete structures and especially concrete bridges. The starting-point of this project is the functional requirements that a structure must meet during its service lifetime. In this thesis an improved and more realistic approach have been developed based on the commonly used probability-based reliability methods, but taking into account the spatial variability of concrete properties. It can provide not only the probability of failure of the structures over time like other reliability methods, but also can produce useful information as the proportion or percentage of the surface area that show concrete deterioration e.g., the area with initiation of rebar corrosion, cracking, spalling, etc. during the whole period of time. This information can facilitate the repair or maintenance strategies for concrete structures from practical point of view. In addition, a maintenance cost prediction model was developed. Decision making of the optimal maintenance or repair strategy is based on the maintenance cost-based optimisation method. The approach developed in the present research is a successful supplement to the current probability-based reliability analysis method and it is also a good way to apply theoretical models in practice, in which the calculation results can be comparable with the inspections directly. The approach is instantiated successfully in three practical cases. The great importance of finding the right spatial variation of the parameters is shown in this thesis so as to evaluate the structural condition correctly. It is strongly recommend that more tests should be implemented specially in the future for finding the real spatial fluctuation parameters of the structures, e.g., the value of fluctuation scale and the common source of correlation coefficient. The approach is worth of more research and development in the future.Civil Engineering and Geoscience