Quantifying the value of structural health monitoring information with measurement bias impacts in the framework of dynamic Bayesian Network

Structural Health Monitoring (SHM) information contributes substantially to Structural Integrity Management (SIM), which can be achieved through reducing epistemic uncertainty and/or enriching decision-making alternatives. However, measurement uncertainties in terms of random error and measurement bias, and the degradation of monitoring performance typically exist. These factors negatively affect the contributions of an SHM system in the context of SIM. To efficiently quantify the Value of Information (VoI) of the SHM system, and to investigate the effects of the influencing factors on the VoIs, this work performs VoI analyses within the computational framework of Dynamic Bayesian Network (DBN). In this framework, Risk-Based Inspection (RBI) planning is used as the prior decision scenario, and two maintenance strategies considering SHM information are proposed as the pre-posterior decision scenario. To demonstrate the significance of taking into account measurement bias and monitoring performance deterioration, two scenarios, considering and ignoring these two influencing factors, are taken into consideration. Finally, the main purpose is demonstrated with a case study associated with optimizing the inspection and maintenance strategy for welded joints subjected to fatigue loading.