PMFA1

from 2012, although by 2015 no orders had been placed. VVER‐TOI was expected to be 20percent cheaper and could be built in 40 months.First concrete was poured for the reactors sited in Russia from 2008–10 and these reactors arewhere substantive experience exists. The major reported incident was the collapse of the steelstructures for a containment build at the Leningrad site in 2011. It was only in 2014 that firstreports of delays emerged and by 2015, all four reactors were 3–4 years late. However, a January2015 report from Russia’s Audit Chamber seemed to put the blame squarely on shortage offunds. 172 Whether there are other construction issues is difficult to tell. Two reactors using anolder design at the Rostov site were ordered at about the same time as the AES‐2006s; one ofthese was completed on time and the other appears close to schedule. It may be that this indicatesmore deep‐seated issues at Novovoronezh and Leningrad than just shortage of capital.The record of AES‐2006 seems somewhat better than that of EPR and AP1000, but the lack ofdetailed information on the AES‐2006 projects and the lack of transparency of the regulatorysystem means it is difficult to draw strong conclusions on the buildability of the AES‐2006compared to AP1000 and EPR.In terms of delays, the record of AP1000 and EPR appears comparably bad. The delays at theChinese AP1000 and EPR sites are similar to each other, and the delays at the U.S. AP1000 sitesare similar to those incurred at a comparable stage for EPR. Both designs have suffered a seriousdesign issue that has delayed construction. The instrumentation & control system for EPR causeddelays because of problems persuading the regulators of its adequacy. The reactor coolant pumpissue was somewhat different, with the problem being that the pump simply did not meet thedesign standard.However, the pattern of construction problems seems somewhat different. For EPR, there was asuccession of quality problems with sitework at both European sites, while the AP1000 sufferedserious quality issues at the module suppliers’ facilities, especially those of CB&I.The promise that Generation III+ designs would be simpler and therefore easier to build appearsnot to have been fulfilled. Real costs have increased significantly compared to their predecessorssuggesting the attempt to reduce complexity was not a success. The attempt to reduce siteworkby shifting the workload to factories through modularized design also does not seem to have hadthe desired effect, and seems to simply have shifted the quality issues from site to modulefactories.1. Standardization has not happened, and all three designs have seen significant andcontinuing design changes. The reality may be that nuclear technology is simply notmature enough to standardize yet and there is still a continuing flow of design changesdriven by experience of operating plants and technical change that it would be foolish toignore. The rate of ordering may also be too low for standardization to be feasible. Ifvendors are receiving only a handful of orders per decade, it seems to make little sense to172 NIW, “Auditor Report Illuminates Rosatom’s Financial Challenges”, 23 January 2015.Mycle Schneider, Antony Froggatt et al. World Nuclear Industry Status Report 2015 64