Will it be alright on the night? Probably. We now know enough about the likely implications of the Y2k problem for the power industry to be cautiously optimistic. In recent months a large body of hard data has been built up internationally, based on investigations of real plants, real components and real systems, which suggests the problem is certainly manageable. Although the finding of resources to fully address the issue in time remains an area of concern for certain countries, the likelihood is that the catastrophists will have a disappointing New Millennium’s Eve.

The body charged with handling the Y2k issue for the US electric power industry is NERC (North American Electric Reliability Council), which has a very useful Y2k web site (www.nerc.com/y2k). According to its most recent status report (issued 11 January), “With more than 44 per cent of mission-critical components tested through November 30, 1998, findings continue to indicate that transition through critical Year 2000 rollover dates is expected to have minimal impact on electric systems operations in North America. Only a small percentage of components tested indicate problems with Y2k date manipulations. The types of impacts found thus far include such errors as incorrect dates in event logs or displays, but do not appear to affect the ability to keep generators and power delivery facilities in service and electricity supplied to customers.”

NERC says the industry is close to the target of all mission-critical facilities being Y2k ready by 30 June 1999. Also encouraging is the degree of participation in the US Y2k effort, which “increased dramatically in the fourth quarter 1998 and exceeds 98% of the electrical systems in the United States and Canada.”

On the nuclear front “No facility has found a Y2k problem that would have prevented safety systems from shutting down a plant…Thus Y2k problems in nuclear facilities do not represent a public health and safety issue.”

Still regarded as a critical issue is the ability to test voice and data communications. These are crucial to power systems, but are usually provided by external suppliers. However the telecommunications industry has given reassurances on its Y2k readiness and demonstrations with power plants are planned shortly. Distribution systems, involving about 3000 organisations in the USA and Canada, are also regarded as critical by NERC because they are on the front line of delivery to customers and “are essentially radial in design and have fewer options that can be used to correct in real time for a failure”. But, reassuringly “results to date indicate that distribution systems are generally the least dependent on electronics and computers and are the least susceptible to Y2k anomalies.”

Nevertheless, despite the expected minimal impacts of Y2k, the US industry is also preparing contingency plans – just in case – and these plans are to be ready by the end of June 1999.

In addition the US power industry will be conducting two practice drills, one on 9 April 1999 (the 99 th day of 1999) and the other on 8-9 September 1999, ie during the rollover to 9-9-99. The latter is one of several other dates that may cause problems for older software (in addition to the familiar millennium bug, the potential inability of older software using two digits for the years to correctly recognise 2000). The problem with 9-9-99 is that in some software 99 and 9999 were used as “end-of-file” markers. Another date to watch is 29 February 2000 which may not be recognised as a leap year in some software (as is 22 August 1999 when the week counter of the Global Positioning System will roll over to zero).

So much for Y2k readiness in the USA, what about elsewhere? ABB, which in 1998 spent $100 million on Y2k and this year expects to spend $150 million, has about 1000 people at plants around the world working on the issue and has built up a large body of experience. Products with microprocessors have been tested to the very strict BSI Y2k standards and the results posted on an internet database (www.abb.ch/abbgroup). According to Klaus Ragaller, ABB Y2k Task Force Leader, “Over 90 per cent of the tested products were found to be compliant, or do not use date functionality. The remaining products that do contain some sort of non-compliance can either be corrected with a patch or replaced by newer versions.” And even when there is non-compliance the kinds of problems likely to result do not look very severe on the whole.

The fact that there are now only ten months to go has seen a sharpening of focus in countries that seemed to be lagging in their Y2k readiness, for example Germany and China. Even Italy now has a national Y2k committee, although its first meeting has only just been held. Russia remains a concern, but there are fewer digital systems to worry about and the country is no stranger to infrastructure failure, so the millennium bug may not make much difference.

But there is very little time and no room for complacency. The Y2k problem is unique in the potential for a variety of failures to occur at precisely the same time, not just in the power sector but in a variety of other essential elements of the infrastructure, resulting in a cascading effect. History shows this kind of multiple failure is often the root cause of serious accidents. So the message must be to keep up the momentum and leave no chip untested.