Assessing Industry Trends In Risk-based Asset Management Practices
Ujjwal Bharadwaj and Polyvios Polyviou
TWI Ltd, Granta Park Great Abington, Cambridge, CB21 6AL, UK
Paper presented at ISOPE 23rd International Offshore and Polar Engineering Conference, 30 June - 5 July 2013, Anchorage, Alaska
The use of risk based approaches in inspection and maintenance planning is increasing as they are seen as cost effective means to maintain availability, reliability and safety standards in plant operations. Taking cognizance of the growing uptake of risk-based practices in asset integrity management, TWI sought to assess trends in risk-based practices by soliciting practitioners’ views through an online questionnaire-based survey. This paper reports the key findings of the survey that will be of interest to a variety of stakeholders in sectors such as oil and gas, power generation and petrochemical.
This paper presents results from a questionnaire based survey carried out by TWI in 2010 to assess trends in risk-based approaches to asset management. The survey reveals some interesting features in the uptake of risk-based practices in industry and provides insights into the current and potential usage of risk based principles in asset integrity management.
The paper starts with an introduction to risk-based approaches followed by some relevant surveys and projects carried out in the past. The features of the survey itself are then discussed before moving on to the results from the survey and some key observations from the survey results.
The conclusions from the survey reflect the opinions of the authors and not necessarily those of others at TWI. They are also by no means the only conclusions that could be drawn from the survey statistics. It is endeavored to present as much raw data as possible so that the reader can draw their own conclusions. It may also be noted that due to constraints it not possible to present all the survey results in this paper. Comparisons have been drawn with some relevant surveys conducted in the past where data has been made available in order to identify industry trends in risk-based asset management practices.
1.1 Risk based approaches
Inspection of plant and machinery has traditionally been mainly based on prescriptive industry practices. However, increased operational experience and a greater appreciation of hazards is leading some parts of industry to adopt a more informed approach to inspection planning, targeting the inspection required to maintain the risk of failure within tolerable level or to reduce it to as low as reasonably practicable.
Risk based approaches are gaining currency as industry looks for efficient and relatively flexible approaches to managing their plant and equipment. When applied to inspection and maintenance of industrial assets, risk based approaches differ from other approaches in the assessment of failure in its wider context and ramification, and the prioritization of inspection or maintenance based on such risk assessment. Simply put, a risk assessment involves combining the likelihood of failure with the potential consequences of such failure. Once a risk assessment is carried out, resources can be targeted to mitigate risks to acceptable or tolerable levels. Apart from the risk assessment itself, the benefits from the structured thought processes involved or leading up to a risk assessment are significant.
1.2 Previous relevant surveys and projects
In 2001, as part of a Joint Industry Project (JIP), TWI Ltd carried out a questionnaire-based survey to gain insights into existing trends in plant life management and the needs of plant operators ( Speck and Iravani, 2002). The survey was sent out to both sponsors and non-sponsors of the JIP and 91 completed questionnaires were received. The survey indicated that the benefits of Risk-Based Inspection (RBI) and Risk-Based Maintenance (RBM) optimization are recognized by different industry sectors particularly the Oil and Gas sector. However, the 2001 survey also highlighted the lack of established and documented uniform RBI/RBM guidance for application throughout industry sectors.
Since 2001, a number of standards pertaining to risk-based practices have had new editions released: API 581( API, 2008), API 580 ( API, 2009), EEMUA 206 ( EEMUA, 2006) and PAS 55 ( BSI, 2008), to name a few. The UK's Health and Safety Executive (HSE) have published a number of reports such as the one on best practice guide for RBI as part of plant integrity management ( Wintle et al., 2001), the one on the management of equipment containing hazardous fluids or pressure ( Wintle et al., 2006) (and the more recent ( Wintle et al., 2012)). Furthermore, the European Commission's project “Risk Based Inspection and Maintenance Procedures for European Industry” (RIMAP) to develop risk-based inspection and maintenance procedures for the European industry got under way and a final report was published; HSE published a report on the implications of the RIMAP approach ( Shepherd, 2005). TWI was involved in a number of research projects that required the application of risk-based principles to Operations and Maintenance (O&M) decision-making and to different asset types ranging from offshore wind farms to ships- ( Bharadwaj and Wintle, 2011); ( Bharadwaj et al., 2011);( Bharadwaj et al., 2007).
1.3 Related developments
The European Fitness-for-Service Network (FITNET) reported the results of a survey on ‘current application and future requirements for European Fitness-for-Service (FFS) technology’ (Filiou et al., 2003) ( Hadley and Kocak, 2008). While RBI is used to prioritize equipment for targeted inspection, FFS is usually applied to a single component with respect to a specific failure mode resulting from some operating damage mechanism. RBI may be used to decide the frequency, location and sampling size of an inspection. Once in-service damage is detected, FFS is applied to evaluate whether the equipment is safe to continue operating.
In 2007, following the release of the new API/ASME joint FFS standard (API/ASME, 2007), TWI carried out an industry survey ( Holtam et al., 2011) to assess the implications of such developments on plant life management practices and to capture FFS trends across a range of industry sectors. Whilst the survey provided insights into current FFS activities as well as how Safety Regulating Authorities (SRA) view these activities and whether or not they accept the results as the basis for plant integrity management decisions, there was lack of information with regards to developments in RBI approaches and their uptake.
2. The 2010 TWI-conducted survey on risk based integrity management practices in industry
2.1 Motivation for the recent survey
The application of formal risk based approaches in industry is relatively new. The last decade saw rapid developments in the usage of risk based principles to plant operation and maintenance in a variety of industry sectors. Apart from keeping up-to-date, industries often find it useful to benchmark their activities against current industry norms and also to learn from practices in other industry sectors. Taking cognizance of such industry needs, it was felt in 2010 that it was time to again assess trends in risk-based asset management practices in industry.
2.2 Features of the Survey: Design, distribution and the return rate
The survey was designed to meet the objectives described above and contained two types of questions:
- Closed-ended (i.e. the respondent is asked to choose from a list of possible answers)
- Open-ended (i.e. the respondent is free to select from a list of answers or write their own answer)
While ‘closed-ended’ questions are useful to carry out quantitative analysis of the responses, ‘open-ended’ questions often provide valuable inputs and provide qualitative information. Thus, a ‘free text’ box was provided at the end of certain questions in which the respondent could make additional comments. ‘Ranking questions’ were also included in the survey to enable respondents to rank a list of choices provided. Finally, some questions had a logical algorithm such that the next question depended on the respondent’s answer to a particular question.
The survey was designed to enable respondents to complete it within approximately ten minutes and could be saved at any time to be completed at a later date. To encourage maximum participation all respondents were given access to the results.
The survey was distributed via e-mail containing a brief explanation and a link to the online survey. The participants were mostly industry practitioners who have interacted with TWI in the past by way of
- purchasing risk-based inspection (RBI) or fitness for service (FFS) software,
- attendance at relevant training sessions and conferences organized by TWI, and
- participating in previous surveys.
TWI is a membership based research and technology organization and the survey was also sent to contacts in its membership database in the Oil and Gas, Power, Renewable Energy, Chemical/Petrochemical and other industry sectors.
Given that the uptake of risk-based practices is increasing across several industry sectors, the survey was not aimed at any specific industry sector, or, indeed, any specific geographical location. However, as discussed later, there is a bias in favor of the Oil and Gas industry sector which unsurprisingly reflects TWI’s involvement in this particular sector.
The total number of completed surveys received was 169, which compares well with other similar surveys - 91 in the 2001 TWI JIP survey, 68 in the 2008 FITNET survey, and 197 in the 2008 TWI FFS survey (Speck and Iravani, 2002; Filiou et al,2003; Holtam, 2008). A survey was deemed to have been completed if it was submitted with responses to more than 50% of the questions.
All respondents to the survey were asked to indicate their personal level of plant management experience; it is pleasing to note that the survey has captured opinions from professionals with a wide range of experience from those who have none or minor experience to those who have spent more than 20 years in the field (Fig. 1). The fact that more than half of the respondents (55.7%) have more than 10 years direct plant management experience provides a high degree of confidence in the results of the survey.
Fig. 1. Plant management experience of survey respondents
2.3 Analysis of results from the survey
The questions asked in the survey can be categorized as:
- General questions aimed at gaining information about the respondent and the respondent’s company.
- Questions aimed at assessing the prevalence of RBI/RBM practices.
- Questions aimed at assessing the acceptance of RBI/RBM practices.
- Questions aimed at evaluating the implementation of RBI/RBM practices.
- Questions aimed at assessing the use of RBI/RBM software.
There is some overlap in the categories listed above. Implementation of RBI/RMB practices, for example, may involve the use of specialized software. However, given the importance of software in RBI/RBM assessments, such questions were grouped separately. In responses, where an overlap was likely to occur, the replies have been normalized. For example, in response to the question on which sector does the respondent’s company primarily operate in, to accommodate the fact that a company could have presence in a number of industry sectors leading to the net percentage being more than 100%, a normalization procedure has been carried out to make the answer more meaningful.
2.4 Responses to general questions
The first few questions of the survey focused on gathering general information about the respondents. For example, respondents were asked to list the region of the world in which their company primarily operated, as shown in Fig. 2.
Fig. 2. Primary regions of operation
Almost one third of the respondents (31.3%) described their company as operating worldwide. Other major categories were Europe (21.3%), Asia-Pacific (18.1%) and the Middle East (12.5%). Only 4.4% of respondents classify themselves as belonging to companies primarily in the North America. However, it may be simplistic to interpret this literally; there are respondents from companies that have a substantial presence in North America and who have classified themselves as belonging to companies operating world-wide. Nevertheless, the geographical spread of respondents is very similar to the results that the 2001 survey reported (Speck and Iravani, 2002).
Respondents were also asked to specify which industry sector their company primarily operated in. More than half of the respondents are based in the broad context of the Oil and Gas sector, with the majority of them primarily operating in the Oil and Gas production. However, since multiple selections were permitted in this question, the results have been normalized by the total number of responses and shown in Fig. 3.
Fig. 3. Primary sectors of operation
The highest proportion of respondents came from the upstream Oil and Gas sector, which is perhaps indicative of TWI’s membership demographic. Sectors in ‘Other’ category included: consultancy, fabrication, aerospace, automotive, construction, mineral extraction and ship building and repairing. In answer to a question about the nature of the company the respondents work for, the majority of respondents (46.4%) came from the producer, operator or manufacturer category. Engineering service contractors were the next biggest category (34.8%); see Fig. 4. Among the responses were participants working for engineering insurers, industry organizations and Safety Regulating Authorities (SRA).
Fig. 4: Survey respondents' company sector and business function (Note: Not shown in the figure is one respondent from SRA associated with Oil and Gas refining.)
2.5 Prevalence of RBI/RBM practices
More than three quarters of all respondents (77.6%) have in their professional capacity been involved in the implementation of RBI/RBM; this aspect in combination with the profile of the respondents’ years of experience in plant management indicates the value of this survey in for defining trends in risk-based asset management practices in industry. In fact, an overwhelming majority of respondents (67.0%) rated themselves more than 5 (on a scale of 1 to 10 where 10 indicated ‘know a lot’ and 1 indicated ‘know nothing’) on their knowledge of RBI/RBM. The percentage fell to 63.9% when queried about their professional involvement in the implementation of FFS procedures, which might be attributed to the fact that the questionnaire was mainly targeted at RBI/RBM professionals. It also reflects the common elements between RBI/RBM and FFS implementations, with the same professionals involved in both to some extent.
Survey participants were also asked to choose among a list of risk and hazard evaluation methods. The results show that about 86 respondents use or intend to use RBI/RBM assessment methods. Hazard and Operability Study (HAZOP) and Failure Modes and Effects Criticality Analysis (FMECA) are the second and third most widely used methods. Fig. 5 shows the breakdown of the usage of risk and hazard evaluation methods as reflected through the respective number of responses.
Fig. 5. Usage of risk and hazard evaluation methods
RBI/RBM is mainly applied to pressure vessels and process piping, followed by application to shell and tube exchangers, and, storage tanks. The number of responses according to the type of asset application is depicted in Fig. 6. This pattern of usage or potential usage of RBI/RBM is consistent with the 2001 survey (Speck and Iravani, 2002).
Fig. 6. Applications of RBI/RBM assessment
A question eliciting respondents’ confidence in qualitative, quantitative and semi-quantitative RBI/RBM techniques and the perceived effort (in terms of time) required in their application offers more insight into the uptake of these techniques. A substantial number of the respondents list qualitative and semi-quantitative techniques in RBI/RBM when it comes to past/future expected usage. The uptake of these techniques is to a large extent influenced by the level of precision in the results required and the resources (in terms of time and money) available. A factor affecting the use of quantitative techniques may also be the requirement of precise inputs which may not be easily available. In terms of accuracy, according to the survey results, quantitative techniques, unsurprisingly, are rated highly relatively by the respondents. However, semi-quantitative techniques are deemed to be reasonably or very accurate by almost 80% of the respondents as shown in Fig. 7.
Fig. 7. Accuracy of different types of RBI/RBM assessments
Although, a lot depends on the context in which these techniques are applied, it seems that semi-quantitative techniques that invariably require some expert input are quite common. This was also the case in the survey conducted in 2001 when an overwhelming majority of respondents mentioned that had used, use and intend to use again semi-quantitative RBI/RBM. However, these figures do not necessarily mean that semi-quantitative techniques are used in the place of quantitative techniques. In terms of time required to conduct Quantitative, Semi-quantitative or Qualitative RBI/RBM, most found Quantitative assessments to be very time consuming and Qualitative ones to be reasonably fast.
2.6 Acceptance of RBI/RBM assessments
It is clear that without support from regulatory bodies the benefits of RBI assessment tools are limited. Responses relating to the acceptance of RBI/RBM by various industry stakeholdersshow that almost 55% of the respondents indicated that their company can obtain exemption from Safety Regulatory Authorities’ (SRA) prescriptive inspection/maintenance rules. This percentage is substantially increased compared to the 20% recorded in the 2001 survey indicating the increasing acceptance of risk based assessments vis-à-vis time or rule based assessments.
Fig. 8. Acceptance of RBI/RBM (Oil and Gas sector)
An overwhelming majority (83.0%) responded positively to the question on whether their SRA accepted RBI/RBM as an alternative basis for determining inspection/maintenance intervals; an example of the break-up for the Oil and Gas sector is shown in Fig. 8. This percentage is substantially higher than the 57% reported in this category in the 2001 survey (Speck and Iravani, 2002), which indicates the increasing levels of acceptance of RBI/RBM assessments. This result is also indicative of how much more widespread and better understood the quality and potential benefits of RBI/RBM techniques currently are.
The survey aimed also to establish which published RBI/RBM procedures were being used most frequently by asking respondents to choose from the following list:
- API Publication 581 – Risk-based inspection, Base Resource Document
- API Publication 580 - Recommended Practice for Risk-based inspection
- ASME CRTD Vol. 41- Risk-based methods for equipment life management
- ASME PCC-3 - Inspection planning using risk-based methods
- RIMAP- Risk-based inspection and maintenance procedures for European industry
- EEMUA Pub No 206 - RBI: A guide to effective use of the RBI process
- Respondent’s company does not currently use any particular published RBI procedure/methodology.
API 580 (68%), API 581 (64%), EEMUA RBI guide (15%) and ASME CRTD Vol.41 (12%) were the most widely used procedures among the respondents, as shown in Fig. 9.
Fig. 9. RBI/RBM methodologies/procedures used
This result may be attributed to the pre-dominance of respondents from the Oil and Gas sector where these procedures are most likely to be used. In the ‘other’ category (7 respondents), which was incorporated as an ‘open text’ box within the question, BS7910, API579, the Australian standard AS3788 (for pressure vessels), RPG 201 (DNV: RBI for offshore topsides static equipment) and some in-house procedures were mentioned.
2.7 Implementation of RBI/RBM
RBI takes considerable resources to develop, implement and maintain. Key stages in RBI such as damage mechanism assessment, risk audit, evaluation of risk mitigation options and inspection planning often require multi-disciplinary inputs covering a range of competencies from experienced plant engineers, as suggested by Ablitt and Speck (Ablitt and Speck, 2005).
In response to a question on how convinced respondents were that an RBI/RBM program was/would be beneficial to their company, an overwhelming majority (90.5%) were ‘reasonably convinced’ or ‘very convinced’ of the potential benefits. It is very interesting to notice that in 2001 the equivalent percentage was only 60% of all respondents (Speck and Iravani, 2002), which highlights the substantial change in acceptance levels and confidence surrounding RBI/RBM programs between 2001 and 2010. Fig. 10, for example, shows the level of confidence in gaining benefits from the implementation of RBI/RBM programs as indicated by respondents operating within the wider Oil and Gas sector.
Fig. 10. Confidence in RBI/RBM programs (Oil and Gas sector)
Furthermore, respondents were asked to rate various factors for implementing an RBI/RBM program. Respondents rated on a scale of 1 to 10, where 10 indicated ‘Extremely significant’ and 1 indicated ‘Not significant at all’, the significance of a given list of reasons:
- Extending the interval between inspection/maintenance
- Improving the overall safety of critical plant
- Classification of plant in terms of potential for environmental damage
- Reducing the duration of an inspection/maintenance outage/shutdown
- Classification of plant in terms of potential for personnel injury
- Reducing inspection/maintenance costs
- Classification of plant in terms of potential business interruption loss
- Producing targeted inspection plans to a structured methodology
According to the survey results, the three highest average scores were for ‘improving the overall safety of critical plant’ (8.50), ‘producing targeted inspection plans to a structured methodology’ (7.78), and ‘classification of plant in terms of potential for personnel injury’ (7.47).
On being asked how they perceived the results from RBI/RBM programs, about 90% of all respondents (among those whose company had previously undertaken a RBI/RBM assessment), indicated that the results of their RBI/RBM program had 'Met expectations' or 'Exceeded the expectations' of their company.
Valuable information is also provided by the survey in regards to the key factors for successful implementation of RBI/RBM programs. Respondents were asked to select from a list of options:
- Appointing a suitable assessment team leader
- Having the 'right' people in the assessment team
- Quality of the historical inspection/maintenance information
- Ease of understanding the RBI/RBM analysis methodology
- Speed of undertaking RBI/RBM assessments
- Reliability of the RBI/RBM analysis methodology
- Tracking decisions and actions arising from RBI/RBM assessments
Having the ‘right’ people in the assessment team is deemed to be the most important factor (with an average rating of 9), while the next most highly rated factors are having a suitable team leader (8.5) and tracking decisions/actions arising from RBI/RBM assessments (8.4).
Figure 11 shows who is responsible for RBI/RBM as indicated by the respondents
Fig. 11. Responsibility for RBI/RBM (overall response from respondents)
2.8 The use of RBI/RBM implementation software
The survey also aimed to capture the respondents’ experience of using RBI/RBM software packages. Most of the respondents (76%) stated that they use software packages for RBI/RBM procedures, with the majority of the software being developed externally (61%). In terms of which attributes of the software were more valued, user-friendliness of the software (with an average rating of 8.3) came first, followed by software based on well-known or published models/methodologies (8.0) and efficient integration into existing management systems (7.9).
In reply to the question on what additional features the respondents would like to see in their RBI/RBM software, a number of interesting points were identified. Some of them are: the ability to integrate with other software systems and databases, having more transparent algorithms, more modules conducting remaining life calculations for other damage mechanisms and other equipment such as heat exchangers, and finally the ability to link with engineering drawings.
3. Key observations from the survey replies
The following key points from the survey results have a bearing on the future uptake of risk based approaches:
- If data from ‘intention to use’ is any indication as to the future direction of RBI/RBM practices, then emerging usages include Pumps/Turbines/Compressors, Structures, and Transportation pipelines.
- Semi-quantitative methods are valued as being reasonably accurate by an overwhelming majority of the respondents. Purely quantitative techniques score more on precision, however they are not always required or feasible to use.
- Strategic Regulating Authorities are increasingly accepting risk based approaches in lieu of the more conventional rule based approach.
- API 581 and API 580 remain the basis for many RBI/RBM applications.
- Although respondents gave almost same weight to a number of possible reasons for implementing RBI/RBM program, ‘Improving overall safety’ and ‘Producing targeted inspection plans to a structured methodology’ stand out.
- Almost 35% of the respondents conduct RBI/RBM in conjunction with external consultants; about 45% view independent (third party) evaluation of their RBI/RBM programs as ‘reasonably important’.
- A substantial number of respondents use externally developed software for RBI/RBM.
- What constitutes a good RBI/RBM software program? The weights given to the factors listed in the question are almost the same. However, in response to the question on what additional features would respondents want in their RBI/RBM software, the following emerge:
- Link to isometric plans or drawings
- Links to maintenance and inspection databases
- Ability to assess more damage mechanisms, different consequences, and the root cause of damage/ failure
- Capability to be a web-based system
- Application to tanks, heat exchangers and piping systems
- Ability to be integrated with CMMS
- Ability to be Integrated with FFS
- Probabilistic assessment capability
- Transparent algorithms
- Open source software.
A review of leading RBI/RBM software is beyond the remit of this paper. However, it is encouraging to note that such features that are seen as desirable by practitioners are already being incorporated in the newer versions of software packages offered by a number of companies. Some of these features are already offered by TWI’s RBI implementation software RISKWISE( TWI, 2012).
The benefits or risk-based methods for asset management are recognized by different industry sectors. The paper has shown some interesting factors in the uptake of risk-based practices in industry. It has confirmed some commonly held beliefs, but it has also revealed a few surprising insights into the usage of risk-based principles in asset integrity management.
The survey shows an increase in risk-based practices in non-conventional industry sectors such nuclear and renewable energy, and also plant applications such as shell and tube exchangers, storage tanks, structures and transportation pipelines. The survey shows an increasing acceptance of risk-based practices by SRAs. It also identifies popular standards used and the common rationales behind the use of RBI/RBM programs. Finally, the survey gives an indication of the key factors in the development of RBI/RBM software.
This paper presents main results and observations from the TWI questionnaire-based survey carried out in order to assess trends in risk-based practices in industry. The authors hope that the work reported here will provide useful input to risk-based asset management approaches being developed and implemented in a wide range of industries.
This paper reports results from a survey funded by TWI Ltd to assess trends in risk based asset management practices. The survey reported in this paper builds on previous surveys carried out in TWI. In particular, the paper has benefitted from suggestions from numerous colleagues, particularly from C.H. Holtam.
- API 2008. API RP 581 Risk Based Inspection Technology. USA: API.
- API 2009. API Recommended Practice 580 Risk Based Inspection. USA: API Publications.
- Bharadwaj, U R., Silberschmidt, V V. and Wintle, J B (2011). Risk based optimisation of spares inventory management. Advances in Production Engineering & Management, 6, 173.
- Bharadwaj, U R, Speck, J B and Ablitt, C J (2007). A practical approach to risk based assessment and maintenance optimisation of offshore wind farms. 26th International conference on Offshore Mechanics and Arctic Engineering (OMAE). San Diego, California, USA.: ASME.
- Bharadwaj, U R and Wintle, J B (2011). Risk-Based Optimization of Inspection Planning in Ships. Journal of Ship Production and Design, 27, 111-117.
- BSI 2008. PAS 55 -1: 2008 Asset Management - Part 1: Specification for the optimized management of physical assets. UK: BSI.
- EEMUA 2006. 206 Risk Based Inspection: A Guide To Effective Use Of The RBI Process UK: EEMUA.
- Hadley, I and Kocak, M (2008). Overview of the European FITNET Fitness-for-Service Procedure. ASME Conference Proceedings, 2008, 447-455.
- Holtam, C M, Baxter, D P, Ashcroft, I A. and Thomson, R C (2011). A Survey of Fitness-for-Service Trends in Industry. Journal of Pressure Vessel Technology, 133, 014001-10.
- Shepherd, B (2005). Safety implications of European risk based inspection and maintenance methodology. HSE Research Reports. UK: Prepared by Mitsui Babcock Technology for the Health and Safety Executive, UK.
- Speck, J B and Iravani, A. Industry survey of risk based life management practices and their relationship to fitness-for-service assessment. ASME 2002 Pressure Vessels and Piping Conference (PVP2002), 2002 Vancouver, BC, Canada. ASME.
- TWI. 2012. RISKWISE software for Risk Based Inspection (RBI) / Risk Based Maintenance (RBM) software for Oil, Gas & Chemical Plant and Power Plant [Online]. Available: http://www.twisoftware.com/riskwise.
- Wintle, J, Johnston, C, Miles, J. and McGrath, B (2012). Management of ageing: A framework for nuclear chemical facilities. HSE Research Reports. Prepared by TWI for the Health and Safety Executive, UK.
- Wintle, J B, Kenzie, B W, Amphlett, G J .and Smalley, S (2001). Best practice for risk based inspection as a part of plant integrity management. Contract Research Reports. UK: Prepared by TWI and Royal & SunAlliance Engineering for the Health and Safety Executive.
- Wintle, J B, Moore, P, Henry, N, Smalley, S and Amphlett, G. (2006). Plant ageing: Management of equipment containing hazardous fluids or pressure. HSE Research Reports UK: Prepared by TWI Ltd, ABB Engineering Services, SCS (INTL) Ltd and Allianz Cornhill Engineering for the Health and Safety Executive.