Paper presented at ESIA 10 - Engineering Structural Integrity Assessment: Present goals - future challenges. Manchester, UK. 19 - 20 May 2009.
For pressure vessels containing non-corrosive, clean and dry products under stable and benign conditions, there may not be any degradation mechanisms affecting the vessel materials over a long time. Taking account of the low risk of failure, it may be possible to justify a longer interval between shutdowns for internal examination. Even so, it is important that this judgement is made carefully and correctly, and that adequate safeguards are in place so appropriate action may be taken if the expected conditions change. This is the topic of a recent report by TWI commissioned by the UK Health and Safety Executive (HSE), which has outlined a six-step approach for assessing and justifying internal examination requirements for high hazard pressure vessels. The work has been developed partly through consultation with leading UK petrochemical companies and inspection bodies that are proactive in applying risk-based inspection best practice.
Experience from UK industry consultation
In order to find out how leading companies operating at high hazard sites in the UK were approaching the issue of internal examination of process vessels, TWI made a series of visits during 2008. The visits included companies from across the UK that were known to have a progressive approach to risk based inspection and operations, involving assessment of equipment in the manufacture of a wide range of products and chemicals. As a result of the visits it was found that in the UK there is an increasing amount of pressure equipment where written schemes of examination (WSE) are specifying extended intervals between internal examinations, subject to a periodic review. Regular and ongoing review of the decision not to inspect inside a vessel may mean that the interval between internal examinations can be extended for a considerable period, although in practice, intervals beyond 12 years are uncommon. In some cases where a justification to extend internal examination intervals has been made, the limitation on the period between outages is due to the restrictions of the process (such as residue build-up or reduced efficiency) instead of the requirement for inspection.
The industrial visits allowed TWI to establish the approach that was being used to implement risk based inspection (RBI) and to draw up written schemes of examination in each case. These discussions helped establish the six step process for justifying extended internal examination intervals (Fig.1). The approach begins with the RBI team assessing the conditions to be free from active degradation during service.
Best practice in risk-based inspection and management
The experts that make up the RBI team play a crucial role in ensuring sound judgements when setting the written schemes and other supporting measures designed to manage the risk of failure. In a collective sense these measures can be considered as risk-based management (RBM), which uses the same RBI approach but applies it to the whole integrity management system. In most situations, the team should include those with competency in the following disciplines:
- Inspection and NDT
- Materials, corrosion and welding (including failure mechanisms)
- Structural design and integrity
- Production processes
- Failure consequences and safety management
- Optional roles may also include production planning, and finance.
Where the RBI team is not a user-inspectorate, it should include someone from the client company fulfilling one or more of the above roles, usually from production or operations. The third party may contribute the specialist knowledge on degradation mechanisms and inspection techniques to assess the likelihood of failure, while the company provides the design, operations and maintenance input to determine the consequences of failure. It is these competencies combined which enable assessment of the level of risk. Competencies for all experts in the RBI (or RBM) team should be demonstrated by proven experience and usually by having suitable technical qualifications or certification. Some individuals may cover more than one competence, and the RBI team may be as few as four individuals, or perhaps as many as 15 to include all the necessary expertise.
Individuals bring their own valuable experience to the team, such that when one of the 'experts' is replaced (through retirement, changing jobs or contractors etc) the RBI assessment may need to be reviewed in the light of the different level of knowledge now represented. When the experts on the RBI team have less experience of the equipment in question (for example, being new to a company, or a third party), it may be preferable to be less ambitious in justifying extending the internal examination intervals (or verifying an already extended interval in the light of changes) than a more experienced team may propose.
The team should meet regularly and assist the Competent Person to draw up the written scheme of examination. Meetings might be planned for before and after a shutdown, or more regularly; each month for example. After a thorough inspection has been carried out or the decision not inspect has been periodically reviewed, the written scheme is considered again by the competent person and the RBM team, and revised for the next inspection. It is at this stage that an extended interval may be included into the new written scheme.
Risk-based inspection is now a well established method for many. However, it is important not to be complacent about the potential pit-falls that can result in a poor quality assessment that falls short of the best practice given in a number of published guides. [2-4]
The first pit-fall is inconsistency. Different RBI teams, be they from different sections of a single company, or from different companies, can give very different results from their RBI assessments due to their different levels of experience and rigor throughout the process. Provided all approaches have conservative assumptions, some scatter does not necessarily matter, but operators should be aware that RBI practices will differ. Following best practice guidance and good training can help.
When RBI teams lack sufficient independence from their company's management then corporate financial pressure may skew the results of an RBI assessment towards particular findings. In a similar way, an RBI may fail to assess all the potential damage mechanisms or risk factors if the RBI team has a 'closed mind' and assesses only what it expects to find, and has insufficient imagination or thoroughness to consider less likely, but potentially dangerous, scenarios. An independent consultant to chair or be part of the team is a possible solution to this.
Sometimes, when third party inspectorates carry out RBI assessments for their clients, information on previous operation or inspection history may not have been made available to them. Where the 3rd party is chosen by a company's accountants (not engineers) primarily on the basis of cost, when the 3rd party used changes frequently thus preventing any continuity to be built up, it is extremely difficult for best quality RBI to be expected.
Finally, a pit-fall to obtaining the best RBI assessment is when there is a lack of safeguards to check that it actually conforms to best practice. Best practice guidance is available, and quality can be improved if measures are in place to ensure it is followed, and/or to audit the RBI process. For extended intervals it is important that these safeguards continue to ensure consistent quality RBI assessments over a considerable period of time, through regular review of the RBI assessment at set frequencies.
Absence of degradation
Equipment for which it might be considered to extend the internal examination interval would primarily be that in 'clean non-corrosive service'. This means equipment that is not expected to suffer from active internal degradation mechanisms over time in normal service or from transient events within the design basis. It is equipment where the product is continuously contained such that water or other impurities cannot reside in the system, where stresses are below code fatigue thresholds, and with high quality and documented welding.
Where any active degradation (such as wall thinning, stress corrosion or fatigue cracking) might occur, regular internal examination is the main method to ensure on-going structural integrity, and these kinds of equipment would be outside the scope for justification of extending internal inspection intervals.
Evidence for the absence of active degradation will be supported by the RBI assessment, in conjunction with previous inspection records over the duration of the equipment's lifetime. This evidence can be difficult to obtain for third party inspectors when the contracts may be changed every turnaround. Further evidence may come from other equipment operating under equivalent service.
The kinds of chemicals under consideration are mainly the range of organic chemical products, which in their pure forms tend to be non-corrosive. Many products that are considered corrosive are stored under conditions where they do not corrode, i.e. very pure chemicals stored within vessels made of suitable containment materials. In these 'non-corrosive systems', where there are no expected degradation mechanisms, the same approach can also be applied. However, the threat of corrosion from even minor contamination with water or other impurities may be significantly higher for these corrosive products, such as in the case of concentrated acids. These types of products therefore need careful consideration when using RBI to set inspection strategies.
If there are no identified corrosion mechanisms and nothing other than pure product within the operational pressure and temperature ranges was ever contained in the vessels, then high confidence can be taken that no environmental degradation will occur and internal examination will not find anything. However, for a robust justification to be made, we must ask what happens if something changes or goes wrong?
The most likely contamination of a clean product is from water. This can be from the production process, leaks in steam heating or water cooling systems, moisture ingress from air as the product is removed from storage, from steam cleaning during outage, or possibly as a result of degradation of the product. For this reason it is unlikely that atmospheric storage tanks would be suitable for extended internal inspection intervals, as these may well be exposed to moisture ingress. The risk from atmospheric water entering the system is that it will contain oxygen or salt, which can result in rusting, corrosion, cracking or pitting of ferrous equipment.
It is also possible that understanding of new degradation mechanisms can come to light for products normally assumed to be non-corrosive. An example is the US experience of ethanol stress corrosion cracking of steel. Keeping up to date with world knowledge along with regular review of the RBI assumptions may avoid ignorance of potential risks.
Justifying extending internal inspection intervals
An overview of the six step approach that can lead to sound decisions on internal examination intervals is summarised in Figure 1. A WSE that does not specify periodic internal examination may be justified, provided the technical basis for the decision is sound, and adequate safeguards such as non-invasive inspection are employed. It is important that a robust management policy is in place to review the justification periodically as well as in response to any change to expected conditions. A company culture that is proactive to reporting changes, and good communications between the operations and integrity management teams, are seen as being essential to sustain the justification. This must be supported by a comprehensive information management system to manage the operational history of equipment.
This section and the next form the main technical rationale for extending interval examination intervals, and the differences between justifications and safeguards can be subtle. As an example, the ability to carry out non-invasive inspection is not in itself a justification not to carry out internal examination, however, given that a justification has been made to avoid an internal examination, non-invasive inspection provides a useful safeguard to allow information to be gathered to support that. A possible strategy is using a stepwise approach that progressively extends the intervals for internal examination as confidence is built over equipment's lifetime.
Form RBI team
- Include people with the all the required knowledge, experience and competencies to carry out Risk Based Inspection (RBI)
No active internal degradation
- Analysis of all possible damage mechanisms
- Combination of process fluids and containment material is stable and inert
- Good quality (documented) fabrication
- Operating/inspection history of vessel
Justify WSE requirements
- Set written scheme of examination (WSE) with review date
- Set operating/process limits for policy validity within a Risk Based Management (RBM) framework
- Reduce the risk as low as reasonably practical (ALARP)
Adequate safeguards and compensating measures
- Non-invasive inspection and maintenance
- In-service monitoring of process, pressure and temperature etc
- Periodic reviews of written scheme and inspection policy
- Opportunistic examinations policy
Management of change policy
- Good knowledge of RBI process settings
- Reactive to equipment modification or changes in product
- Live process i.e. fast response to upset or abnormal conditions
Organisational and management aspects
- Peer review and audit of management and operating procedures
- Maintaining awareness from world experience
- Sustaining the competencies of the RBI team
Figure. 1 - Overview of the process to justify and support extended internal examination intervals
Usually, the justification for setting extended inspection intervals is based upon a risk based inspection (RBI) assessment to demonstrate the absence of active degradation mechanisms. The RBI assessment also includes considerations of the consequences of failure when setting the internal inspection strategy. The RBI assessment must outline clear validity limits where the assumption of absence of degradation is still valid, and when it is not. The justification should be conditional on being reviewed periodically as part of the review of the written scheme of examination, and in the event of expected conditions occurring or new knowledge being available, by there being a robust management policy in place to review the justification in light of these. For some high hazard situations, where the scale and consequences of failure are very high, moving to a completely non-invasive inspection scheme is unlikely ever to be tenable and there may be grounds for continuing to carry out some internal examination, even though no active degradation mechanisms are identified.
Risks of carrying out internal examination
There are inherent risks in carrying out internal examination, and undertaking an examination might actually increase a previously low risk of vessel failure if suitable precautions are not taken. On the other hand, a skilled internal examination might give better quality inspection data than using a non-invasive technique.
The risks of internal examination include both that to the vessel (e.g. from depressurisation, steam cleaning, grit blasting, or internal scaffolding), and those to the inspectors themselves (such as from working in confined spaces or at height, or from residue fumes). The poor visibility and restricted access, combined with the required protective equipment for the operator, may impair internal inspection results so they provide only limited value. The justification to extend examination intervals is therefore also influenced by the risks, difficulty and costs of undertaking the examination.
Safeguards when extending internal examination intervals
Given that it is justified that an internal examination is going to be required on an infrequent basis, there are actions which can also be taken to compensate for the inherent side benefits of regular internal inspection, and to safeguard against any potential disadvantages of not examining the inside of vessels and detecting unanticipated conditions outside the assumptions made. One of the main methods is non-invasive inspection.
Information about the thickness and the condition of the internal surfaces of vessels can under some circumstances be obtained without entering the vessel using non-invasive techniques. These techniques include NDT methods applied from the outside of the vessel, such as ultrasonic testing, shearography and thermography. Non-invasive techniques also include those where endoscopes or laser probes can be used to 'visually' examine the inside of a vessel when inserted from a remote location through a hand-hole for instrument access.
Non-invasive (also called non-intrusive) inspection (NII) is not always cost saving; the methods can be slower and more expensive to apply than internal visual inspection. However, it may still be preferable to internal examination insofar as it avoids the need for human access and possibly emptying the vessel. When applying NII it is important to know where to look, recognising that welds (and in particular, weld roots) are usually the most critical locations to inspect, but when the caps have been ground flush they can be hard to locate from the outside. External examination using ultrasonics can assess wall thickness loss or hydrogen induced cracking (HIC), but should not be relied upon to find internal surface breaking stress corrosion cracking or pitting.
Process and product monitoring
Without internal examination, in-service monitoring of product composition and purity gives confidence that the product has not degraded or been contaminated, and therefore that the clean, non-corrosive conditions expected are actually being maintained. Deviations in concentrations or chemistry from monitoring process fluids can help to show when process-upsets have happened, or detect the by-products of corrosion. It is important to recognise, however, that sensors intended simply to measure product quality may not be detecting the correct constituents, nor the correct accuracy required for an integrity assurance argument. The power of using in-service monitoring in an integrity safety case is in knowing what properties and impurities to look for, how to detect them, and interpret that data.
Devices such as pressure relief valves and thermocouples provide important evidence to confirm whether service pressure and temperature conditions are within the expected operating window. Usually safety valves are maintained and reset every time equipment is shut down for internal examination. However, regardless of justifying longer inspection intervals, these safety devices will still need to be inspected in accordance with their own technical requirements, and it may be these that limit the shutdown period given in the WSE. Provided that another protective device is in place during operation, some safety valves can be removed and tested (or replaced) while the equipment is on-line. Doubling up on pressure relief valves is becoming more common in order to do just this, and to reduce the amount of work required during a shutdown.
Whenever a vessel is to be taken out of service (for reasons other than to carry out an inspection), the opportunity should be taken to make an internal visual examination of the vessel if possible.
It is also important to ensure that bolts, seals and other ancillary components can still be inspected, tightened, calibrated and overhauled as necessary during the extended inspection interval, in a similar fashion as the safety valves.
A further safeguard is to provide measures that will mitigate the hazard arising from a failure, in the worst case scenario identified. This may be done using a leak-before-break integrity argument or possibly by providing secondary containment to contain any leakage. An analysis of the consequences of a failure is needed to allow Duty Holders to be prepared for any action they may need to take.
Management of change
If there is no internal examination for long periods, particular care is needed when feed product or process conditions are changed, or when changes are made to other parts of the plant, as these could alter the composition and impurities of the products. These changes might be planned and expected changes to operating condition, or they could be a result of unexpected upset conditions, which might be transient or permanent. The impact of such changes on the susceptibility to degradation needs to be considered. Modern companies manage both planned and unexpected changes to their high hazard equipment through a robust Management of Change policy.
A Management of Change committee should include relevant experts who can discuss the effects of any changes or incidents on the RBI strategy. In leading companies, unexpected incidents requiring the full Management of Change committee to meet may occur only infrequently, but the policy gives confidence that the RBI approach remains robust under all eventualities.
It is necessary to initially have adequate systems in place to warn the appropriate people of any relevant change to the established parameters, such as level gauges, pH alarms, processing pressure and temperature gauges. The warning should alert someone who has the authority and understanding of the implications of the change to convene the RBM team (or a suitable sub-set of the team) together. Good communications between the operations and engineering teams is a key requirement. Systems then need to be in place to evaluate the importance of the reported events on the inspection policy and decide what action is necessary. A responsive system that can act quickly is key, but this can be more difficult where the full RBI team may be spread across a number of companies or sites, and communications may be poor.
The large amount of equipment at a refinery, chemical or petro-chemical plant means that managing all the data about the equipment is done using computers. Specialist software is commercially available to help with this, and also with managing the RBI and planning the inspection of each item. The information management system ideally needs a way of recording operational history, including for example the number of start-ups/shutdowns, pressurisations, temperature transients and changes to contained product. Trends or anomalies in the conditions of the equipment or product can indicate ahead of time where issues might occur.
Handling, validating and maintaining data is a costly and time consuming business, but it does allow ongoing fact-based decision making instead of relying on opinion and conjecture. Companies that have made this investment are more likely to benefit from the better quality and accessibility of their data making it easier to justify potentially longer intervals between internal examinations.
Appraising the RBI approach
When RBI is being used to set inspection strategy for pressure vessels with high consequences of failure, it is good practice for companies to undertake an independent peer review and audit of the management of their RBI assessment process. This might be an internal review within a large company by those with suitable expertise but not involved directly in the RBI or plant in question. An alternative is for an external review, possibly by an independent competent body or consultancy organisation. There are audit tools available in HSE documents that can help provide a structure for such audits.[1,3] The intention of the peer review would be to examine the stages of the processes that the RBI team has gone go through to reach its decisions, and the factors that were considered in reaching the inspection decision. It is not intended that the peer review effectively does the whole job a second time.
The team of experts comprising the RBI team needs to maintain its awareness of on-going developments in their respective fields. A review of how the members of the team have been selected, how their competencies have been demonstrated, and what information has been gathered for the RBI assessment and so forth can add rigour to the whole process.
When an RBI assessment predicts no internal degradation, then a written scheme of examination that specifies an extended period for internal examination may be justified provided the technical basis for the decision is robust and sufficient non-invasive inspection is achievable or other adequate safeguards are employed. In such cases it is important to review the justification periodically, as well as in response to any change to expected conditions, and to take opportunities for examination when they arise for other reasons. It should be noted there may be some very high hazard situations where a completely non-invasive inspection approach is unlikely ever to be entirely tenable, because of the potential failure consequences and need for defence in depth.
A pro-active company culture to reporting change and good communications between the operations and integrity management teams, supported by a comprehensive information management system of the operational history of equipment, are seen as being particularly important to sustain the justification. Maintaining the depth of experience and level of competency required to make sound RBI assessments is a major challenge for companies. While the large multi-national companies are able to share experience rapidly across their operating sites with similar processes and equipment, smaller organisations can access the same level of expertise and knowledge through appropriate consultancy and professional activity. The need for staying up to date with new developments was widely recognized. There are important roles which regulators, industry bodies, professional institutions and international organizations can take in improving the dissemination of experience and research.
- Moore, P. and Wintle, J., Extending the internal examination intervals of high hazard process plant, HSE Research Report, to be published 2009.
- Engineering Equipment & Materials Users' Association, EEMUA 206 Risk-based inspection - a guide to effective use of the RBI process, January 2006.
- Wintle, J., Kenzie, B., Amphlett, G., and Smalley, S., HSE CRR 363, Best practice for risk-based inspection as a part of plant integrity management, 2001.
- American Petroleum Institute, API 580, Risk-based inspection: downstream segment, American Petroleum Institute recommended practice, 2002.
- Det Norske Veritas, recommended practise DNV RP G103, Non-intrusive inspection, October 2007.