This case study illustrates one of the earliest applications of fitness for purpose technology. More than 20 years later, TWI continues to offer this service to Member Companies, using the latest assessment techniques.
Following disclosures to the press of falsification of a small number of radiographs, the Aleyaska Pipeline company ordered an audit of all radiographs on the 30,000 welds of the 800-mile long 48-inch diameter pipeline. This line now carries about one million barrels of oil daily from the fields on the Alaskan North Slope - Prudhoe Bay - to Valdez.
The audit showed that about 4,000 welds contained imperfections, which should not have been permitted according to the workmanship standards stipulated in the pipeline code, AP1 1104. This finding did not relate to the falsification but was due to incorrect interpretation of the radiographs during construction.
After the audit, the question arose, did these imperfections prejudice the service performance of the pipeline? This was studied by TWI in collaboration with Cranfield Institute of Technology using methods developed at TWI and now incorporated in BS7910. Modes of failure were studied which could have been influenced by the imperfections - brittle and ductile fracture, fatigue, corrosion assisted fatigue and stress corrosion cracking.
The investigation showed that all but the most severe imperfections found during the audit were acceptable on a fitness for purpose basis, and the pipeline company submitted a case to the US Government seeking waivers so that the line could enter service without expensive repair. This plea was finally accepted. However, time was against the operators. For about half its length the pipeline transverses perma-frost and it is only possible to work in this zone in winter when the surface layer is fully frozen. The operators decided that they could not afford the risk that their case for waivers would prove unsuccessful and that they would then have missed this seasonal window. They therefore proceeded with the repairs regardless of the possible outcome of their plea. The repair programme, almost entirely for harmless flaws, cost approximately $50M. One repair alone is reputed to have cost $2M - this was to a weld in a river crossing and involving the construction of a cofferdam. The repair in this case simply involved grinding off the cap of the weld, which contained a group of totally innocuous pores, no repair weld metal being required.
Although, on this occasion, TWI was unable to save its Member's money because of the timing, the project did show the tremendous potential for cost saving by the use of fitness for purpose assessments. Nowadays the regulators should be able to reach decisions much more quickly because of the availability of standardised documented procedures.
After the project, the Engineering Manager of the pipeline company wrote expressing 'appreciation of the excellent work done recently by TWI relating to the fracture mechanics and associated studies of pipe line girth welds'.