PSI - Issue 64

Fadi Oudah et al. / Procedia Structural Integrity 64 (2024) 1983–1989 Fadi Oudah/ Structural Integrity Procedia 00 (2019) 000 – 000

1988

6

7. Practical Recommendations A. Need for calibrating spatial-temporal fields . Spatial-temporal fields simulate random patterns of mechanical deterioration over time. Several formulations exist in literature although they have not been applied in structural engineering practice. Practical application of spatial-temporal fields to model the random deterioration patterns of freeze-thaw actions and corrosion of reinforcing steel should be based on a proper calibration. The calibration should ideally be based on existing structures as opposed to accelerated lab tests. B. Need for efficient on-site approach for determining the correlation length . The correlation length is often selected by conducting a limited sensitivity analysis to recommend values that produce conservative results. This approach is not realistic and will likely misrepresent the reliability of the examined system. It is recommended to develop user-friendly methods to assess the correlation length from field measurements (destructive or non-destructive) of the examined RC structure to represent the three categories of spatial variabilities discussed in Section 4. C. Need for calibrating a professional factor for the capacity model determined using RFE . When RFE is used to develop a resistance model for reliability analysis, a professional factor random variable should be introduced to address the uncertainty in the method of analysis. Several recommendations for the statistics of the professional factor exist in literature for conventional sectional analysis for concrete subjected to flexural, shear, and torsion, but no information exist for reliable statistics when the capacity is assessed using RFE. D. Need for calibrating the random fields such that to produce exact realizations at the location of in-situ testing . Coring concrete cylinders is commonly used in the evaluation of existing structures to assess the compressive strength. The number and locations of concrete cores are determined based on the condition of the structure and following standard procedures including ASTM E122-17 and ACI 214R-11. There is a need for calibrating random fields to produce exact realizations at the locations of the concrete cores and provide random patterns of strength elsewhere within the space of the structure. This approach would improve the accuracy of the resistance model when used in reliability analysis. E. Need for exploring procedures to improve the efficiency of RFE simulations . For RFE simulations to be widely adopted in structural engineering consultancy, the simulations need to be efficient. Efficiency in this context refers to 1) efficiency in developing the RFE model, and 2) efficiency in determining the minimum number of models needed to establish a representative resistance model when RFE is used in reliability analysis. The former item can be addressed by modeling the spatial variability in critical zones within the structure that will likely govern the capacity (concept of influence zone introduced by Khorramian et al. (2023a) for geotechnical engineering applications). The latter can be addressed by developing a framework for reliably enriching the RFE capacity histogram when used to fit a distribution type to develop the resistance model for reliability analysis. The enrichment should be accompanied by a stopping criterion. Summary and Conclusions Random finite element (RFE) can be used to assess the capacity of reinforced concrete structures experiencing spatial variability due to the intrinsic variability in the concrete mechanical properties and the deterioration over time due to freeze-thaw damage, alkali aggregate reactivity (AAR), corrosion, etc. The random field properties can be estimated based on destructive or non-destructive field testing of existing structures. The application of RFE in the evaluation of in-service reinforced concrete (RC) structures using a reliability-based approach is generally academic, while limited effort has been put to apply it in real-life engineering consultancy. This paper documented three recent uses of RFE for the reliability evaluation of existing RC structures experiencing spatial variability due to environmental deteriorations. General observations and recommendations for the practical use of RFE in fast-paced structural engineering consultancy are made. It is evident that the application of this method will grow in future for the evaluation of existing RC structures as new software such as RF-DYNA is developed and used in practice.