PSI - Issue 62

Sebastian Thöns et al. / Procedia Structural Integrity 62 (2024) 259–267 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

265

7

^LJƐƚĞŵ ,ĂnjĂƌĚƐ ĂŶĚ ĚĞƚĞƌŝŽƌĂƚŝŽŶ ŵŽĚĞůƐ ŽŵƉŽŶĞŶƚƐ ƐƚĂƚĞƐ ŵŽĚĞůƐ ^LJƐƚĞŵ ƐƚĂƚĞƐ ŵŽĚĞůƐ

^LJƐƚĞŵ ŵŽĚĞů

ĞŶĞĨŝƚ ŵŽĚĞů ŽƐƚ ŵŽĚĞů ŽŶƐĞƋƵĞŶĐĞ ŵŽĚĞů ŝƐĐŽƵŶƚ ƌĂƚŝŽ

hƚŝůŝƚLJ ŵŽĚĞů

DŽĚĞů ďĂƐŝƐ

ĐƚŝŽŶ ŵŽĚĞů ĐƚŝŽŶ ŝŵƉůĞŵĞŶƚĂƚŝŽŶ ƵŶĐĞƌƚĂŝŶƚŝĞƐ ĐƚŝŽŶ ĐŽƐƚƐ /ŶĨŽƌŵĂƚŝŽŶ ŵŽĚĞů /ŶĨŽƌŵĂƚŝŽŶ ĂĐƋƵŝƌĞŵĞŶƚ ƉƌĞĐŝƐŝŽŶ /ŶĨŽƌŵĂƚŝŽŶ ĐŽƐƚƐ

^LJƐƚĞŵ ĂŶĚ ĐŽŵƉŽŶĞŶƚ ŵĂŶĂŐĞŵĞŶƚ ŵŽĚĞů

/ŶĨŽƌŵĂƚŝŽŶ ŵĂŶĂŐĞŵĞŶƚ ŵŽĚĞů

Fig. 5: Main characteristics of a decision scenario

3.1 Case study: Fatigue monitoring of the Great Belt Bridge A service life extension of the Great Belt Bridge in Denmark is investigated in conjunction with fatigue monitoring in Long, Alcover and Thöns (2021). Here, the monitoring durations and the extension of the service life of an orthotropic steel bridge deck are analysed with a decision theoretical approach. The model basis for this decision scenario encompasses the structural performance assessment and prediction with a probabilistic fatigue model, monitoring data-based fatigue life prediction (including monitoring costs, service life benefits (tolls) and failure consequences). The decision of a service life extension of the bridge deck is determined through maximisation of the expected utility (i.e., maximizing the expected benefits and minimizing the structural risks), and in conjunction with the determination of an optimal fatigue monitoring period. The decision analysis is constrained with the required target reliabilities. The repair of fatigue damages is also modelled. A parametric analysis is also carried out to determine the effect of the target probability, benefit, cost of failure, cost of rehabilitation, cost of monitoring, and discount rate on the posterior utility and the choice of service life. The findings indicate that the strategy of conducting short-term monitoring, specifically for one week every six months, leads to the highest expected utility. Furthermore, the analysis reveals that the target probability constitutes the parameter with the highest influence on the optimal monitoring duration and the service life extension. Here, it should be noted that the service life extension is determined with the accumulated maximum expected utility over the service life within the boundaries of the economic and technical service life limits. 4 Summary and conclusions The service life constraining principle of infrastructures constitutes non-reparability or non-adaptability, attributable to evolving environmental hazards and structural degradation. Such reparability and adaptability are constrained by technical (specifically engineering), economic, and life safety boundaries. The assessment of the technical service life should be initiated by evaluating the feasibility of prevention, the capability for detection, and the potential for repair of damages. The analysis of structural degradation mechanisms necessitates the adoption of probabilistic methodologies, owing to significant inherent uncertainties and hence the necessity for periodic updates based on new data and information. Degradation models should facilitate the prediction of probabilistic estimates of damage size over time, allowing for their integration into the limit state models. Consequently, the structural reliability can be compared against the corresponding target reliability complying with life safety requirements, thereby determining the service life without the need for additional (durability) limit state modelling. Estimating the service life based on structural reliabilities without any observation based updating nor measure based adaptation addresses solely the technical aspects of service life, where deterioration cannot be prevented or repaired. A holistic service life assessment, encompassing economic, technical, and safety considerations, can be achieved