By Taf Powell, International Regulators Forum
Editor’s note: This article is the first in a series of articles to be published during 2008 that will focus on the critical issue of asset integrity. The International Regulators Forum kicks off the series in this issue, presenting findings from a three-year inspection programme. In upcoming issues, The Petroleum Safety Authority (PSA) of Norway, Australia’s National Offshore Petroleum Safety Authority (NOPSA) and the Netherlands’ State Supervision of Mines will all provide their perspectives on how industry can improve asset integrity.
A major programme of inspection by the UK regulator focused on installation integrity on the United Kingdom Continental Shelf (UKCS) reveals the need for major improvement. This article describes some of the significant findings and outlines conclusions arising from a three-year programme of inspections carried out by the UK regulator, the Health and Safety Executive’s (HSE) Offshore Division, focused on asset integrity. Although MODUs represented only 10% of the programme, there are some important lessons relevant to drilling sectors of the offshore industry.
The offshore oil and gas industry on the UKCS is a mature production area. Much of the offshore infrastructure is at or has exceeded its intended design life, and many of the MODUs operating there are aging. During the 1990s, low oil prices and initiatives to reduce costs led to a reduction in the offshore workforce. This in turn led to reductions in levels of maintenance, and, as a result, to an overall decline in the integrity of fabric, structures, equipment and systems.
In 2004, HSE announced the commencement of Key Programme of inspections (KP3) focussed on asset integrity. The main drivers for this inspection programme were rising levels of concern over asset integrity and two fatalities in 2002 resulting directly from integrity failure. The programme focused on the management and maintenance of safety critical elements (SCEs). SCEs are those parts of an installation and its equipment (including computer programmes) that are barriers designed to prevent, control or mitigate major accident hazards (MAH), and the failure of which could cause or contribute substantially to a major accident. A good maintenance management strategy should provide assurance that these barriers will be available when required, that they will operate with the required reliability, and that they will survive incidents long enough to mitigate the hazard maintenance of SCEs. The relationship between MAH, development of SCEs and their maintenance is shown in Figure 1.
The coordinated KP3 programme of inspections used a consistent approach, and data in the inspection reports were collated and analysed by a central team. This analysis identified areas of good and poor practices and enabled benchmarking of the whole UK offshore industry. This has facilitated engagement and been effective in raising the profile of integrity management.
In total, nearly 100 installations from 33 Duty Holders were included in the exercise. The programme included fixed installations, floating production installations and MODUs. For each element inspected, a rating in the form of a colour code – red, amber or green – was assigned. Green indicated that no degradation in the element was found. Amber indicated that HSE thought it appropriate to send a letter to the owner/operator raising issues of concern. If enforcement action was considered, then a red rating was assigned.
Figure 2 shows an extract from the KP3 report. It indicates the overall performance of the industry for elements of company management systems. Each line in the table represents the results for an installation whilst each column represents an industry perspective of an aspect of the maintenance system. The KP3 report and the full version of Figure 2 can be found on HSE’s Offshore Oil and Gas website at www.hse.gov.uk/offshore/information.htm.
KP3 Finding Summary
The programme revealed important cross-industry findings that can be grouped under the following key headings:
Leadership and control of major hazard
• There is a poor understanding across the industry of the potential impact of degraded non-safety critical systems on the control of a major accident.
• The role of asset integrity and concept of barriers in major hazard risk control is not well understood.
• Companies need better key indicators of performance available at the most senior management levels to inform decision-making. Many management monitoring systems tend to be overly biased to occupational risk data at the expense of major hazard precursors.
• Many senior managers are not making adequate use of integrity management data and are not giving ongoing maintenance sufficient priority.
• There is wide variation in the condition of hardware integrity across the offshore industry dependent on installation type and design, CAPEX costs and subsequent investment.
• The main hydrocarbon boundary appears reasonably well controlled, but ancillary hydrocarbon infrastructure such as valves continue to be in decline.
• Primary structural integrity is reasonably well controlled.
• Insufficient full loop testing is carried out on SCEs, resulting in reduced levels of reliability.
• There were often considerable variations in performance of management systems between assets in the same company as well as between companies.
• Significant improvement in maintenance systems could be achieved without major capital expenditure by better planning, improved training and clear statement of performance standards in testing and maintenance routines.
• The industry is not effectively sharing good and best practice.
• Cross-organisational learning processes and mechanisms to secure corporate memory need to be improved.
• Companies need to work better with Verifiers, using their collective skills and knowledge to aid improvement.
• Audit and review arrangements are not being used effectively to deliver organisational learning and continuous improvement.
• In some companies, the decline in integrity performance that started following declining oil prices has not been effectively addressed.
• Declining standards in hardware is having an adverse impact on morale in the workforce.
• Skills shortages and long lead times for delivery of materials and equipment are limiting the industry’s ability to achieve rapid improvements.
In broad terms, the asset integrity management of MODUs was found to be better than for fixed installations, which in turn was found to be better than for floating production installations.
The 10 MODUs inspected performed better than the other types of installation in all elements reviewed other than the maintenance of SCEs and maintenance basics, for which the performance was comparable. Here temporary refuge (TR) HVAC, TR doors and deluge in particular gave significant cause for concern. These were all key issues identified at the time of the Piper Alpha disaster.
For the maintenance of SCEs, the issues were related to:
• Lack of a formal maintenance strategy;
• Poor implementation of maintenance change requests;
• Weak links between performance standards and work orders;
• Generic performance standards that were not measurable and hence difficult to demonstrate that they have been met.
Acceptance criteria that provide confirmation that performance standards are met were often absent. As a result, onshore management have been unable to monitor whether their SCEs actually meet their performance standards. An example of measurable criteria is valve maximum closure times. An example of not being specific is where the same performance standard is used across all of a company’s installations despite there being differences in the actual systems on the installations.
There is evidence that the offshore workforce do not understand links between the safety case, MAH analysis, identification of SCEs and development of their performance standards. The workforce is the last and critical line of defence against the occurrence of many incidents. It is therefore essential that they fully understand that the equipment they work with provide barriers against MAH.
Good practice found in relation to maintenance of SCEs included:
• Ensuring a clear link to the performance standard on the work order.
• Ensuring that the post-maintenance system tests relate to the performance standard requirements and are clear and equipment-specific.
• The offshore workforce and management being provided with training regarding the roles SCEs have in preventing, controlling or mitigating MAH.
The better MODU performance compared with other installation types in relation to the overall physical state of the plant might be due to:
• Being less complex and generally smaller in size.
• Having no major hydrocarbon processing systems resulting in generally less complex maintenance tasks.
• Having a market value for resale (unlike fixed production installations), which encourages good upkeep of rig equipment.
• Not continuously drilling, whereas a production installation is continuously producing, meaning that there is often more opportunity to undertake maintenance.
• Clients auditing MODUs extensively before and during hire periods.
• There being a strong commercial incentive to keep rig downtime to a minimum, with revenue being dependent on reliability and readiness rather than on throughput of hydrocarbons.
• Classification requiring regular structured inspections.
• MODU owners comprising smaller, leaner management teams and usually having shorter reporting lines to and closer relationships with senior management than producing companies do.
Potential Causes for poor performance
HSE have determined a number of potential underlying causes for the performance pattern emerging from the programme and have grouped them under issues concerning leadership, the engineering function and learning.
Senior management set priorities between investment in asset maintenance and profit on the basis of health, safety and financial risks. The findings indicate that the priority given to asset maintenance has been too low. Whilst most senior managers currently use information on maintenance performance such as backlogs and deferrals, this provides only a limited picture on SCE status.
To set priorities better, senior managers need to improve their understanding of the safety and business risks arising from operating with degraded SCEs and safety-related equipment. This may require a simplification of the reporting arrangements, including the development of focused performance indicators, for backlogs, deferrals, corrective maintenance, SCE and asset integrity performance, etc.
It is not an easy task to determine the optimum balance between maintenance and profit. Senior managers need to make decisions based on good, reliable data and informed technical opinion. The difficulty, of course, is that data and opinion contain uncertainty and bias.
Revenue from an asset is likely to be reasonably well known, whereas the actual or potential costs due to skill shortage or diversion of resources to workarounds are less easy to predict accurately. The actual or potential costs associated with declining morale caused by an apparently deteriorating asset and increasing workarounds are even more difficult to estimate. However, probably there is most uncertainty in quantifying the increased safety risks arising from a combination of degraded SCEs, declining morale and reduced risk control.
Financial analysts provide forecasts and cost benchmarking figures for the industry. Senior managers can draw on this to support decision-making. Over time this becomes an iterative process that provides sustained pressure to reduce costs. Significantly, however, there is no similar iterative process to benchmark and drive down risks.
HSE’s advice over uncertainty in risk management is to be conservative. Senior managers need to understand and allow for uncertainty in all data. HSE Information Sheet 3/2006 (Guidance on Risk Assessment for Offshore Installations) discusses uncertainty in risk assessment and the principle of using sensitivity analysis.
Senior managers should not use this type of data alone in managing investment in an asset but should aim to maintain or improve standards using qualitative as well as quantitative assessments, giving appropriate weight to technical arguments advanced by the engineering function.
Where uncertainty clouds the figures, senior managers could simply adopt clear objectives on asset integrity and put in place a process to:
• Eliminate maintenance backlogs and deferrals and reduce corrective maintenance.
• Eliminate or minimise operating with degraded SCEs.
• Minimise or eliminate ORAs.
It is possible to get sharp indicators for each of these, and a focus on this would promote balance in the iterative process of cost reduction.
The engineering function
In the last 20 years, there has been a huge increase in the volume of data available. There are better analysis tools, better statistics and quicker ways to manipulate data. This improves reliability predictions for operations such as HPHT wells, extended-reach drilling programmes, and Reliability Centred Maintenance (RCM) programmes. This has allowed the industry to extend operations safely.
On the other hand, increasing use of data manipulation can give a false sense of safety assurance if the underpinning engineering knowledge is not there. There continues to be a shortage of skilled engineers in the offshore and other industries. This has resulted in increasing work demands on engineers, reducing support for technical committees, less time available for peer review, and less time to build the experience required to make sound judgements.
These trends can lead to decision-making being over-reliant on data rather than on good engineering standards and judgement. The latter ought to provide a backstop against degraded SCEs and safety-related equipment and structure. Therefore, HSE believes that there should be a strong engineering reporting line to board level.
KP3 has demonstrated that there is considerable variation in the performance of maintenance management systems and delivery of appropriate standards, across the offshore industry and often in the same company. This is true even for the management elements which are not a function of the age of the installation or the size of the company. Hence there is plenty of scope for companies operating a number of installations to improve their asset management systems by capturing and embedding learning.
Managing continuous improvement, like managing risk, needs to be formalised if it is to be effective. Improvements can be identified in activities, but it requires a blend of knowledge of the process and the operational context and an independence of mind to challenge the status quo. Furthermore, an improvement climate is intertwined with a risk management climate. Some teams excel in fostering a good risk and improvement culture. The members of such teams share good practice and performance improves. Sharing this good practice between teams and between installations may require formal processes and rely on people being open to change.
Intranets and internets have greatly increased the ability to share information, but sharing is only the first step in learning. Learning requires best practice to be evaluated in the context of the particular company or operation and then to embed it in operational and training systems. Activity reviews, audits and incident learning are processes that can be used to successfully promulgate good practice.
The Offshore Division of HSE (OSD) found that many companies’ audit arrangements are not designed to promote improvement but rather just to ensure compliance with procedures. Figure 3 shows a conceptual model for the audit process that includes a learning element. In safety cases, many companies claim to have better than average SMS performance through continuous improvement, but few have formal management structures to achieve this.
OSD currently uses a template based on the maturity model that categorises the ability of the audit process to promote learning during inspections. The learning principles can be applied to other systems such as maintenance, operations, drilling and incident management. Figure 4 shows characteristics of the incident management process in terms of maturity. Current work indicates that most companies operate around level 3, with some closer to 4.
The ability of a company to capture learning in all management systems can improve performance and balance the potential negative effects from cost reduction.
What needs to be done?
The signs for improvement on the UKCS are positive but must be sustained by senior managers increasing their focus on major hazard control and creating a culture where good practice and lessons learnt are promulgated both throughout an organisation’s operating assets and across the industry as a whole. These are important issues and certainly have relevance to the drilling industry and its management of the major accident risks present everyday in the business of searching for, and developing offshore oil and gas.
Taf Powell is head of the Offshore Division of the Health & Safety Executive.