Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

Integrating Diverse Approaches to Reliability of Engineering Structures

Project Code: 32878
Start date and planned duration: February 2019, 36 months


  • Compare the probability of failure of selected welded structures using both risk-based assessment and fracture-mechanics-based assessment.
  • Demonstrate whether/how the two methods can be integrated.

Project Outline

The project will develop TWI’s capabilities in assessing the reliability of safety-critical welded structures, and integrate two different approaches: Method 1 (risk-based, or ‘top-down’) and Method 2 (fracture-mechanics-based, or ‘bottom-up’). The benefits for industry are expected to be substantial, including: (i) integration of two disparate methods of quantitative reliability assessment, which are currently typically undertaken by different teams; and (ii) an understanding of the probability of failure associated with Engineering Critical Assessment (ECA) carried out using currently-accepted deterministic methods.

The project will focus initially on a single failure mode (brittle fracture under monotonic loading), both because of its practical importance and because of its amenability to modelling using fracture mechanics techniques, both deterministic and probabilistic. Different types of welded structure will be considered, with particular attention paid to:

  • Structural redundancy: for example, fixed offshore structures exhibit a high degree of structural redundancy, so that failure of one member will not lead to failure of the overall structure, whereas non-redundant structures such as pressure vessels may fail catastrophically because of breach of the shell only. Nevertheless, it is standard practice to manage structural integrity of both types of structure using a range of techniques including: ensuring certain levels of materials toughness and strength, post-weld heat-treatment of thick sections, pre-service/in-service inspection and control of stresses/stress concentrations.
  •  The role of residual stress in welded steel joints. Residual stresses associated with both the as-welded (AW) and post-weld heat-treated (PWHT) condition will be considered, along with the implications of rules and standards that mandate application of post-weld heat-treatment for steels above a certain thickness.

 Where possible, fatigue crack growth will also be modelled probabilistically, with a view to linking current risk-based research addressing fatigue with a probabilistic fracture mechanics-based fatigue approach. This is expected to be a more challenging, but equally important, endeavour compared with consideration of brittle fracture, since the uncertainties associated with fatigue crack growth parameters and initial flaw size can lead to even greater uncertainties in fatigue life.

Industry Sectors

Benefits to Industry

The outcomes of the project will be fully documented, transparent and traceable; consequently, it will be possible for industry to apply the methods developed during this project to their own structures. This will increase the accuracy of assessments of the reliability of structures, improving safety, and potentially reducing both fabrication and operating costs.


 Research Board Portal

For more information please email: