Year 2000 glitch presents problem of unprecedented scope for petroleum industry

Offshore installations present a number of technical and safety challenges related to the Year 2000 date change. Operators of such installations must inventory components and control systems and rehabilitate those that are not capable of making the date change at the turn of the century. Shown here is Oryx Energy Co.'s Neptune production spar (center), flanked by J. Ray McDermott SA's DB-50 derrick vessel (left) and Diamond Offshore Drilling Inc.'s Diamond Ocean Lexington drilling rig (right). Photo courtesy of Oryx Energy.

• Year 2000 project timeline [53,661 bytes]
• Three facets of the year 2000 problem [139,253 bytes]

As petroleum firms around the world work, both literally and figuratively, against the clock to make their operations Year 2000 compliant, a number of obstacles remain.

Software portfolios, operational control systems, and facilities must be readied for the date change. These tasks involve a huge testing and remediation workload, and they are complicated by a scarcity of both the resources and the authority needed to perform them.

Complicating matters is the necessity of preparing for the ubiquitous threat of legal liability in the event that a company's lack of preparedness causes problems for customers or suppliers.

The Year 2000 problem reaches well beyond information technology (IT). It is a business viability issue. Critical systems must continue to work during the millennium change in order to ensure that operations, and therefore profits, do not cease.

Time is running out

Only 15 months remain in which to deal with the Year 2000 issue-even less time, in some cases, as data that contain dates in 2000 and beyond can be introduced into computers and other business systems at any time. If, for instance, a company plans major equipment shipments 9 months in advance of a construction project that begins Jan. 1, 2000, dates in 2000 may begin to appear in their systems as early as Mar. 1, 1999.

This is called a company's "date horizon." It is because of this horizon that many companies intend to complete their program this year, leaving next year as a buffer (see diagram, p. 30).

In addition to the by now well-known challenges of the Year 2000 problem, many companies' programs are being complicated by mergers, acquisitions, and joint venture formations that continue to amass in the petroleum industry. Depending on the terms of these deals, their effects on a company's Year 2000 program can be limited in scope or far-reaching.

Immense proportions

The Year 2000 problem is one of the most comprehensive and complex issues ever faced by the industry. It presents a threat to the viability of governments as well as to manufacturing and service companies of all ilks.

It is all but impossible to think of a company or organization that is not affected by this insidious computer glitch.

It is ironic that a problem so all-encompassing stems from a simple abbreviation: the recording of a four-digit year as a two-digit number. Yet this truncation that at one time seemed harmless may, as the year changes from 1999 to 2000, have major effects, such as crashing airline reservation and air traffic control systems, or minor ones, like bringing elevators and escalators to a halt.

In the petroleum industry, its potential effects range from something as simple as the prevention of employees from entering the office parking lot gate by means of a magnetic swipe card to something as costly as bringing a refinery or offshore platform off line. Petroleum firms commonly divide the Year 2000 issue into three main divisions: computer software, control systems, and interactions with outside parties (see chart. p. 31).

Each of the three major divisions of Year 2000 problems is complex, and each requires a different approach. Finding two-digit dates in software programs is mainly an IT function, while finding Year 2000 compliance problems in control systems involves operations personnel and often outside vendors.

Identifying and preparing for all such potential problems is an enterprise-wide project, and one with a scope never before encountered by businesses, say many observers.

"Year 2000 is a gigantic program management, project management, and risk-mitigation issue," said Thomas McAndrew, senior partner in Computer Sciences Corp.'s consulting and systems integration division in Boston.

At a Petroleum Equipment Suppliers Association executive briefing in Houston in February, McAndrew said, "The problem is simple, the solution complex. It's not just an IT issue; it tends to be a very, very real business issue."

There are deadlines to be met, manpower and capital must be "diverted" to fix the problem, and there is a potential for failure and subsequent problems, he says. He calls it the "Year 2000 triple constraint"-resources, time, and business risk.

Royal Dutch/Shell agrees. On its Internet site-a place where many companies are making their Year 2000 plans public-Shell says, "The challenge is not so much a technical one-although, the more work done on the problem, the more complexity seems to arise-but one of resources and fixed deadlines.

"Many Shell people know the excitement and frustration of planning and executing a project, from the modest to the mighty. So try to imagine every business and organization on earth faced with the same issue and all forced to resolve it in precisely the same time frame.

"Quite simply, it is the biggest project in the history of the world."

And, as of Oct. 1, there are only 457 days in which to solve it.

A resource drain

"It is probably the most expensive project in history, too," said Shell.

At a London conference in April, Shell U.K. Ltd.'s director of corporate affairs, John Mills, said, "If we are to 'get it right' in the run-up to the Year 2000, and into the next century, significant resources, in terms of personnel, finance, and management decision-making, need to be devoted to meeting the Year 2000 challenge."

The estimated cost to business and industry of fixing the problem is constantly changing, said McAndrew. Estimates have ranged from $300 billion to $600 billion worldwide (not including litigation costs, tool costs, machine resources, end-user acceptance, and other items).

Litigation costs alone have been estimated at more than $1 trillion worldwide.

Because of the costs involved, the Year 2000 problem is expected to have a material effect on strategic spending for many companies.

As of May this year, British Petroleum Co. plc was planning to spend a total of $150 million on the Year 2000 problem, and ARCO says it will spend about $20 million.

As for Shell, David Trim, a consultant with Shell Information Services' Year 2000 team, said, "I inherited a figure of $100 million for Shell when I arrived, but you can assume it will be more.

"You can't put a figure on it with any degree of certainty," he added. "It could be more than that for Shell in a single country. What is certain is that the figure is constantly being revised upwards."

Suncor Energy agrees. In its 1997 annual report, Suncor said, "The company is unable to estimate at this time the total future costs expected to be incurred that are associated with completion of its Year 2000 compliance project, although costs associated with certain upgrades have already been incurred or included in budgets in the normal course of business."

Regarding the allocation of Year 2000 outlays, Mills said that about 70% of Shell Exploration & Production's budget will go toward ensuring asset integrity, for a very simple reason: "The consequences for the company, and for the country, of a platform failure, for example, would be great. Deferred production costs alone could run well in excess of £1 million/day.

"Similarly, if the St. Fergus gas fractionating plant were to fail, a large part of the North Sea would be unable to operate."

Shell Expro will spend only about £2.5 million on business computing systems. In the downstream sector, that figure will be about £3 million.

Risk

The risks involved in the Year 2000 problem are staggering.

McAndrew separates the potential effects of Year 2000 problem into five categories. In descending order of severity, these are:

• Lights out (ceasing of main business function).
• Legal liabilities.
• Financial problems.
• Regulatory problems.
• Nuisances.

"Every issue and every problem-whether it's facilities (or) IT-have to be looked at to determine what your risk is," said McAndrew. "There is definitely a continuum of risk. For every company, that risk is different."

Petroleum companies are at high risk for experiencing a failure of some sort because they use a broad range of software, some of which was designed in-house, and because they operate literally thousands of devices that contain embedded computer chips.

The higher a company's risk, the greater its need for a comprehensive and consistent Year 2000 program, says McAndrew.

The first thing a company must do is determine its most critical business function. All other operations must then be prioritized, in terms of the risk of failure that they pose.

It is unlikely that any firm will be able to find every single line of software code and every microchip that contains date sensitivity. It is therefore necessary that each company finds a level of business risk it can live with.

This is a matter of weighing the time and resources involved against the risk, says McAndrew. His suggested approach to developing an effective Year 2000 program includes the following steps:

• Create a compliance approach.
• Establish a communication program.
• Develop a plan.
• Enforce a consistent methodology.

A consistent methodology can provide legal protection against problems that might occur if different divisions of a firm use different methodologies. If one division's systems fail while another's remain intact, it will be more difficult for the company to show that it did everything it could to prevent a Year 2000 failure.

In addition to developing a consistent and detailed plan for dealing with the problem, says McAndrew, you need the buy-in of everyone throughout the process, from the top to the bottom, including lawyers, customers, and suppliers.

Clive Mather, head of Shell Information Services, agrees: "ellipseThe message has to go out across the group that this is very important.

"Although the problem was created by the IT industry, its implications are much more widely spread, due to the general use of control electronics. Managers have to assess to what extent their businesses may be at risk, and where getting this wrong may leave them, in terms of a competitive edge."

Integration

The solution to the Year 2000 problem may appear simple: change data from a two-digit year format to a four-digit format. But the complexity of doing this-in terms of technology, cost, and personnel-make it a much larger problem than it seems at first.

The reason for this, says McAndrew, is that computer systems are now very complex and highly integrated.

"Data are no longer self-contained within a company," he said. "(Today), we all survive on shared data.

"In 1960, if somebody pulled the plug, it affected your company. In the Year 2000, if somebody pulls the plug on your machine, it affects not only your company (but also your suppliers and customers).

"So the supply chain, in the data sense of the word, is a big issue (see figure, p. 34 [124,359 bytes]). It is necessary to understand your supply chain, both from a goods perspective and a data perspective."

These interdependency issues must be addressed through a complex analysis of interactions with other businesses, in terms of data, products, and services.

Leland G. Freeman, vice-president of Source Recovery Co., Framingham, Mass., says Year 2000 program managers should include a technique called "dependency modeling" in their plans: "The model should identify, as accurately as possible, all interrelationships between your organization and other organizationsellipseModeling software can help you articulate these relationships and determine how your organization will be affected."¹

Such modeling is crucial to Year 2000 preparedness. McAndrew warns that, even if a company is fully Year 2000 compliant and has done its job 100%, a company with which it does business can cause it to experience a failure.

Compliance definition

The definition of "compliance" is another issue complicating the Year 2000 problem. There is a difference between being Year 2000 "compliant" (meaning all dates are in four-digit year format) and being Year 2000 "capable."

McAndrew says that Microsoft, for example, says its products will be Year 2000 capable, which means that some of its applications will use a two-digit year in conjunction with what is called a "pivot year"-commonly 1950.

Here's how it works: If a two-digit year is entered and its value is between 50 and 99, it is assumed to be in the 20th century. If, on the other hand, a two-digit year is entered and its value is between 00 and 49, it is assumed to be in the 21st century.

This means that Year 2000-capable systems will not recognize dates before 1950 or after 2050.

According to McAndrew, what Year 2000 program managers are finding out about system compliance is that:

• New equipment and software will be Year 2000 compliant.
• Upgrades will be Year 2000 capable.
• Communication between divisions of a firm will be Year 2000 capable.
• External communication will be only "consistent" in its date format.

He suggests that companies should be willing to provide data in the format in which they wish to receive it: "Don't expect to receive four-digit years if you're only willing to provide two-digit years."

Authority problems

Because of the onerous time and manpower requirements involved in the Year 2000 testing and remediation process, one common barrier to on-time completion of Year 2000 programs is a lack of the necessary authority to allocate resources.

Companies are not used to doing enterprise-wide projects such as Year 2000, says McAndrew. As a result, most Year 2000 programs have an accountability vs. authority problem, he says, meaning that program managers do not have near as much authority as they have accountability.

Many Year 2000 program managers don't feel that they have the authority necessary to do whatever is needed to make their programs a success, says McAndrew.

"What we find in large companies is that the program (management) does nothing more than carry out the Year 2000 project: they collect data, they report status, they provide information, but they can't make the decisions.

"You cannot have a successful program unless somebody out there can make the decisions that authorize the expenditures," he said. Otherwise, you're just a clearinghouse.

"For your Year 2000 effort, you need to take each of your initiatives, you need to lay them out, and you need to determine who is the controller and (if they are) involved on a regular basis in helping us manage this project.

"As the time comes closer and closer, you're going to have to make tough decisions," he says.

While many companies are instituting enterprise-wide software systems-such as SAP AG's R/3 or Oracle Corp.'s suite of software products for the energy industry-many of those projects are in progress, and often they are behind schedule, says McAndrew.

He asks: What happens if a company determines that it is not going to make it by the deadline?

"How hard do you think it's going to be for somebody to go to the CEO and the board and get another $2 million allocated to go back and fix the old system-which you're now getting $10 million to replace with SAP-because your SAP project is not going to make it in time?

"Those are the tough decisions that people will have to make between now and the Year 2000. And they are not going to be made by the project manager and IT."

You need to involve a lot more than just IT people, he says. You need people who can handle the necessary internal communications as well.

"Don't let IT drive the solution; let the business problem drive the solution."

BP agrees that the problem is much larger than IT. An internal company memo on the Year 2000 issue stated: "The business knowledge required to validate results often does not exist outside the business unit. While an IT staff can certify that the software code runs properly, it can't answer whether the software still performs the business functions it was built to provide. Only the business unit can certify that."

Year 2000 support must be visible at the highest level of companies, says McAndrew: "Not the CIO, not the CFO, not even the CEO, but it's the CEO and the board of directors that are going to have to answer to (Year 2000 problems)."

"Companies still haven't taken ownership of the problem," he said. "Most people take the attitude: I think I'm in good shape. It's a perceived status, not a fact-based status," he said.

"Part of the problem," according to the BP memo, "is that many CEOs are still in denial as they look for a cheap way out. Even very professional IT departments tend to underestimate the scope of work required."

One company, for example, began work on its year 2000 project in 1995, said BP. At that time, it estimated that it had 30 million lines of code to fix. Two years later, that estimate had doubled to 62 million lines. Among other things, that company discovered that intercompany testing is a massive problem that will cost as much as its internal Year 2000 project work.

Liabilities

Legal issues are a major consideration in the quest for Year 2000 compliance. Many law firms now have departments focused solely on Year 2000 litigation.

The best approach to legal problems is "due diligence," says McAndrew. That is how a company protects itself, because no firm can fix 100% of its potential Year 2000 failures.

In addition to taking the necessary measures to find and solve Year 2000 problems, companies must also take steps to verify the compliance of their suppliers and customers.

McAndrew says a company needs to know not only whether its suppliers and customers are compliant, but also how they achieved compliance. "If they're using a pivot year, you need to know what that year is," he said.

He also warns that, just because a software or hardware vendor says its product is compliant, that doesn't absolve the buyer from having to test it.

There is good news, says McAndrew: "Most companies will be able to hold up if they can show that they went through an organized process to determine where their failures are, remediate the failures, evaluate their business risk, and allocate dollars and that they have been forthcoming in disclosing this information with their suppliers, with their customers, and with stockholders."

He believes 60% of companies in the high-risk category will have some sort of failure in their organization.

Will the world be ready?

The Gartner Group, a consulting firm and provider of Year 2000 solutions, surveyed 15,000 companies and government agencies in 87 countries regarding their Year 2000 preparedness programs. Based on the results, Gartner concluded that most major companies and agencies in most of the developed world are doing relatively well in preparing for the date change.

Two of the surprising laggard countries were world economic powerhouses: Germany and Japan. And there is not much time remaining for firms and agencies in these countries to catch up.

"Everyone is running against the clock," said Steve Rowlands, Year 2000 project manager for CMG U.K. Ltd., London.

Predictions about the effects the Year 2000 problem vary greatly.

Ed Yardeni, chief economist at Deutsche Bank Securities in New York, said "The Year 2000 problem is a very serious threat to the U.S. economy. Indeed, it is bound to disrupt the entire global economy.

"If the disruptions are significant and widespread, then a global recession is possible. Currently, I believe there is a 60% chance of such a worldwide recession, which could last at least 12 months starting in January 2000 and could be at least as severe as the 1973-74 global recession."

Likewise, David L. Babson & Co. Inc., investment counsel, says the bug could have significant economic effects: "According to some estimates, the Herculean efforts involved in correcting this so-called Year 2000 problem may cost trillions of dollars and a half a percentage of national growth in the years 2000 and 2001."

And noted IT expert Peter de Jager estimates that 1% of companies worldwide could go out of business as a result of the bug.

On the other end of the spectrum are those who believe that the problem is so well-recognized and potentially catastrophic that it will be fixed, barring a few minor glitches.

The oil industry, as a general rule, seems to fall into this category. Cooperative efforts undertaken by the industry should help ensure that any Year 2000 failures will be relatively minor and easily repaired or dealt with through contingency planning (see story, p. 32).

What to expect

"Year 2000 failures are now starting to occur on a regular basis," said McAndrew. "We still, at this point in time, have not stepped up to the problem as well as we should have." And this is all industries, and in all segments of industries, he says.

"Companies will start seeing failures, more and more frequently, (beginning) right now," McAndrew predicted. "In this situation, their 'date horizon' is going to come into play."

Many people think the problem won't occur until Jan. 1, 2000, but it's not true.

He works with a company that provides 3-year financial forecasts for investors, so this firm started seeing Year 2000 problems last January. Their systems failed, he said. Their reply was, "Don't worry. We'll have it fixed by the end of March." That could have been avoided, said McAndrew.

According to Don Estes, chief technology officer for 2000 Technology Corp., "The goal of Year 2000 testing cannot be to find all faults in advance. That would require more resources than virtually all organizations have availableellipseThe realistic goal is to reduce the spike to a manageable level by spreading out the period of time in which failures will occur. In other words, the goal is business survival."²

References

1. Freeman, Leland G., "Year/2000 Black Holes: What You Can't See Can Hurt You," Year/2000 Journal, May/ June 1998, p. 75.
2. Estes, Don, "Testing for Unknown Risks: Why Management Must Influence Testing Programs," Year/2000 Journal, July/August 1998, p. 62.

This first of two articles on the Year 2000 problem in the oil industry focuses on the scope of the problem and the potential implications of Year 2000 failures for overall business viability.

Copyright 1998 Oil & Gas Journal. All Rights Reserved.