Asset integrity management for sustainable industrial operations: measuring the performance

Abstract

The management of the integrity of engineering assets such as production plants, facilities and equipment can be a challenging task due to their interrelationships with financial, information, human and intangible assets. In order to ensure effective utilisation and to improve the management of such assets, one has to make balanced and effective decisions based on performance assessments. The main challenge for asset integrity management (AIM) is related to various aspects of its human dimension as apparent on organisational settings and associated cognitive dispensations. In this article, AIM is cascaded down to design, operational and technical integrity management. Furthermore, the performance of physical assets (PA) is discussed in terms of financial, societal and environmental dimensions that deliver sustainability value to the assets' owner. The management of such a sustainable value is vital in managing the overall asset integrity in order to minimise rising stakeholder pressure and achieve balanced performance. The paper provides a review of AIM and a foundation for engineers and managers to analyse the general problem of managing PA' integrity to increase the sustainable value of an asset intensive business in a more holistic way.

Keywords:

• sustainability
• balanced performance
• asset performance
• asset integrity management

Introduction

The concept of sustainability (also referred to as ‘triple bottom line’; Elkington 1997) is an inherently vague (Daly 1990) and compelling concept, which remains difficult to express in concrete and operational terms (Briassoulis 2001, Ratnayake 2010). The ‘triple bottom line’ addresses an expanded spectrum of values and criteria for measuring organisational success in terms of achieving financial, environmental and societal interests in a balanced way. However, when it comes to industrial asset operations, environmental and societal concerns are expressed as health, safety and environment (HSE). Furthermore, nobody can resist the argument that all assets of an industrial organisation should contribute to preserving the quality of the societal and ecological environment for future generations.

However, in practice, lacking a sustainability concept has brought some of the best leading global corporations almost into bankruptcy. One of the best examples of the consequences (in terms of financial, environmental and societal considerations) due to lack of asset integrity (AI) in a broader perspective is the Macondo well blowout in the Gulf of Mexico. A leading oil and gas company, BP, was almost brought to bankruptcy and their Deepwater Horizon accident investigation report reveals that the work force

did not identify any single action or inaction that caused this accident. Rather a complex series of interlinked mechanical failures, human judgment, engineering design, operational implementation and team interfaces came together to allow the initiation and escalation of the accident (BP 2010).

The aforementioned indication reveals that the necessity for a holistic integrity assessment (in terms of ‘triple bottom line’ dimensions) is paramount as usually multiple companies, work teams and circumstances are involved with asset operations. This is the case especially for offshore oil and gas operations, where maintenance, cleaning, operations, drilling, cementing, etc. are carried out by different contractor companies.

However, due to economic recession, globalisation and stakeholder pressure, companies today are searching for ways to utilise every conceivable advantage to keep pace with their competition. Furthermore, the industry in general is under increasing pressure to reduce costs, to maximise returns, to meet tougher performance and production targets and to comply with regulatory requirements on assets (Ouertani et al. 2008). On these grounds, asset-centric industries such as energy, utility and process are looking for opportunities to manage and/or reduce stakeholder pressure as well as the cost of maintaining their assets; to improve the performance and extend the life of those assets; to speed up information analysis and decision-making processes and to gain competitive advantage throughout the asset life cycle.

The management of assets' integrity tends to be broad in scope, covering a wide variety of areas such as asset management (AM), integrity management (IM), AI-related aspects, performance measurement and sustainable performance (triple bottom line vision). Amadi-Echendu et al. (2007) state that understanding the complexity of an evolving field in an industry requires understanding the boundaries of the specific activities contained within that field. Furthermore, the closely related fields broaden the insights and views about the emerging field while providing breeding grounds for learning from each other. Therefore, in developing a definition of asset integrity management (AIM), the authors have utilised the ideas coming from the general field of AM, IM, and also associated AI assessing sectors. The definitions of the terms asset, IM, AI and AIM will be given in later sections.

Although the AI of an organisation plays a vital role, Payne (1994) and Pearson (1995) stated that at that time, 15 years ago, there was a lack of any conceptual or integrated framework to enable organisations to review and improve their integrity themselves. Furthermore, Giacalone and Greenberg (1997) observed that integrity breaches/violations are generally not studied in relation to each other and a patchwork of models and insights has arisen. This makes it difficult to adopt a holistic approach to analyse integrity violations in practice. Moreover, Mohseni (2003) argued that asset managers must ‘be able to justify decisions and policies and assess the impact of each decision upon the key business values against which the asset base is being managed’. Hence, it is vital to have a holistic framework in order to map AIM activities and measure the asset performance and to adopt business strategies and activities that meet the needs of the enterprise and its stakeholders today, while protecting, sustaining and enhancing the human and natural resources that will be needed in the future (also see Deloitte and Touche 1992).

This article provides a comprehensive review of AIM concepts that have been previously published within different disciplines. Then, it proposes a holistic approach for measuring AI performance via gauging to what extent the personnel are sufficiently capable of balancing conflicting stakeholder demands. The approach is aimed at providing dynamic and consistent feedback about the current performance of the assets to provide a basis for improving decision making for managing the AI in a balanced way without compromising long-term benefits for short-term gains. Finally, the manuscript proposes a framework to measure industrial AI performance. The framework is developed based on the findings of previous research carried out for measuring organisational alignment (Ratnayake and Markeset 2010a), technical integrity performance (Ratnayake and Markeset 2010b) and operational integrity performance (Ratnayake and Liyanage 2009).

Reviewing AIM concepts

In this section, the authors provide the reader with a background in AIM by reviewing the existing literature. First, we shall define what we mean by an ‘asset’, and then the AI, IM, AM and AIM are discussed in the context of physical AM.

Asset

Depending on the level of the organisational hierarchy and discipline, the term ‘asset’ carries a different interpretation and usage. For instance, Swanson and Curry (1989) have defined the term ‘asset’ from an information management point of view as ‘a qualified entity that, through its reuse, improves quality, provides a competitive edge and reduces software development and support costs’. When it comes to an infrastructure perspective, the term ‘asset’ is defined as ‘a physical component of a manufacturing, product or service facility which has value, enables services to be provided and has an economic life of more than twelve months’ (IIMM 2002). The British Standards Institution (BSI) has released a Publicly Available Specification (PAS) (see BSI PAS 55-1 and 2 2004), which is applicable to any asset-intensive business, such as energy, transport, manufacturing and utility industries. In PAS-55, a more focused definition of ‘asset’ is used, namely ‘plant, machinery, property, buildings, vehicles and other items and related systems that have a distinct and quantifiable business function or service’.

However, Baskarada et al. (2006) defined an asset as ‘anything of economic value that is owned by an organization’. While this is a more general description of what an asset is, Snitkin (2003) and Koronios et al. (2007) preferred to distinguish between the broad range of assets an enterprise has. They classified assets as tangible assets (liquid assets such as cash or inventories and fixed assets such as buildings, infrastructure, information technology (IT) equipment, machineries, hardware, and product and service equipment) and intangible assets (ITA; designs, knowledge, software, intellectual property and processes).

Based on experience and communications in the oil and gas industry, we choose in this article to adapt the definition proposed by Ouertani et al. (2008), namely

Definition 1 (Asset): ‘An asset is defined as any physical core, acquired (i.e. the organization has either the possession or the custody of the asset) elements of significant value to the organization, which provides and requests services for this organization’.

In Definition 1, the authors assume the physical core (or asset) to be an industrial plant, an O&G production facility, etc. which is comprised in relation to other critical categories of assets: financial, human, information and ITA (see also BSI PAS 55-1 and 2 2004). Thus, an industrial plant or PA under consideration is seen as an element that generates revenue, provides and requests services to/from its user throughout its life cycle, and requires considerable effort and cost in ensuring effective utilisation.

As an example, on an offshore O&G production facility, the flowline system represents the PA and the human assets (HA) are represented by operators, inspectors, inspection planners, maintainers, etc. To perform their job, the personnel rely on the information about the PA from different sources (historical and recent inspection data about corrosion/erosion, technical condition reports, findings reports, etc.). The information asset (IA) assists the HA to control the technical condition and plant health and to make decisions, which prevents leakages and reduces financial as well as HSE losses. Once the technical condition is at an acceptable level, the burden in relation to financial assets (FA) is minimised through lesser insurance payment, less compensation for fatalities, fewer fines due to violation of regulations, fewer production losses due to shutdowns, as well as preventing damage to the production equipment, machines, systems, processes, facility, etc. This also increases the value of the ITA and may result in employee satisfaction, goodwill, reputation and a long run-time for a license to function within the area.

Hence, it is vital to view PA in relation to other critical categories of assets such as financial, human, information and ITA. However, this requires effective and efficient organisation and management processes.

Asset management (AM)

According to Amadi-Echendu et al. (2007), there has been a debate since the 1990s about how to have an appropriate blend of skills and an interdisciplinary approach for handling the complex nature of AM. They argue that the biggest challenge for asset managers are most likely due to the ‘various aspects of human dimension as manifest on organisational settings and associated cognitive dispensations’. Mohseni (2003) backs up this argument and states that effective AM is not only maintenance, capital investment, whole life cost analysis or risk but also involves striking a balance between financial and non-financial metrics.

Engineering aspects of AM

The definitions of AM cover two distinct but important aspects of the management of assets. The first aspect is mainly centred on the engineering aspects in relation to information and communication technology in which assets are managed by monitoring their condition to prevent premature deterioration. Madu (2000), for example suggested that maintenance, reliability and cross-organisation analysis are the key issues in managing equipment asset use, arguing that ‘asset management’ is facilitated by IT software. Galusha (2001), considering IT assets, defined AM as ‘a combination of tools and processes that proactively manage a company's entire asset from a cost, contractual, support, and inventory viewpoint’. A profit equation was described as well where ‘assets+people = profit’, which also highlights the key role of HA among other critical asset categories.

In the report on the ‘21st century asset management’ issued by the American Association of State Highway and Transportation (AASHT 1997), AM is defined as ‘a systematic process of maintaining, upgrading, and operating and PA cost-effectively’. They add that AM is ‘facilitated by tools that help decision-makers perform asset management effectively’. To generate and access quantitative and qualitative data on an organisation's assets, enabling technologies (i.e. computerisation, electronic sensors, robotics, global positioning system (GPS), satellites) are utilised.

Recently, in an article discussing maintenance strategies, determinations of component condition, asset simulation, statistical fault analysis and statistical AM approach (distribution) and life management (transmission) in relation to distribution and transmission network operators, Schneider et al. (2006) describe AM as ‘operating a group of assets over the whole technical life cycle guaranteeing a suitable return and ensuring defined service and security standards’. Similarly, Koronios et al. (2007) argue that

asset management entails preserving the value function of an asset during its life cycle and maintaining it to as designed or near original condition through maintenance, upgrade, and renewal until sustainable retirement of the asset due to end of need or technology refresh.

However, as argued in the AASHT report (AASHT 1997), AM ‘provides an opportunity for both horizontal and vertical integration within an organization’ and thus is not just an engineering tool. AM transcends and includes disciplines such as finance, planning, engineering, personnel and information management.

Total management of PA

The second aspect of AM is centred on the total management of PA in relation to the other critical categories of assets – not only IA but also financial, human and ITA. At the same time, the management systems related to these categories are integrated and managed to inform decision making about those assets (see also BSI PAS 55-1 and 2 2004). For example, McElroy (1999) argued that a focus on effective AM requires an asset decision-making framework to incorporate organisational structures and IT aligned with financial and budgetary considerations. The Federal Highways Authority in the USA developed an AM primer to guide thinking and activities in a systematic way and in order to understand the critical elements of AM (AMP 1999).

Tsang (2002) has suggested that the human dimension is a key issue for the successful management of engineering assets. Complex interactions of skills and resources, PA specificity and the way these assets are managed are discussed in Reed and Defillipi (1990).

However, a more holistic definition of AM is provided in the British Standard PAS 55-1 and 2 (BSI PAS 55-1 and 2 2004). The standard declares that AM is ‘the optimum way of managing assets to achieve a desired and sustainable outcome’ and its objective is

to ensure (and to be able to demonstrate) that assets deliver the required function and level of performance in terms of service or production (output), in a sustainable manner, at an optimum whole life cost without compromising health, safety, environmental performance, or the organization's reputation.

The construct ‘asset management’ has also been defined in a range of different contexts including plant, machinery, property, buildings, vehicles, and other items and related systems that have a distinct and quantifiable business function or service (PAS 55-1 and 2 2004), transport (AMP 1999, McElroy 1999), construction (Vanier 2001), electricity (Morton 1999), chemical engineering (Chopey and Fisher-Rosemount 1999) and irrigation (Malano et al. 1999).

The above definitions and views of AM reflect the general trend towards the importance of AM rather than just asset maintenance to focus on the broader picture of life cycle asset assessment, including strategy execution, risk and performance measurement, health, safety, environment and human factors, see also Townsend (1998), Mitchell et al. (2007), Schuman and Brent (2005) and OECD (2001). However, the focus of this article is the management of PA in relation to other critical assets such as financial, information, human and ITA. From this point of view, this article adopts the definition proposed by Woodhouse (2006), which reflects our point of view of AM.

Definition 2 (Asset management): AM is ‘the set of disciplines, methods, procedures and tools derived from business objectives aimed at optimising the whole life business impact of costs, performance and risk exposures associated with the availability, efficiency, quality, longevity and regulatory/safety/environmental compliance of an organization's assets’.

Integrity management (IM)

The integrity of an organisation is achieved via generating and accessing quantitative and qualitative data from an organisation's assets. Hence, enabling technologies (e.g. computerisation, electronic sensors, robotics, GPS, satellites) are vital prerequisite for enhancing the integrity and for achieving smooth functioning of an industrial organisation (Pearson 1995, LeClair et al. 1998, Kaptein and Wempe 2002). The enabling technologies provide sensible information to satisfy stakeholders' demands (Hill 1990, Williamson 1993, Shaw 1997, Schwartz and Gibb 1999, Kaptein and Wempe 2002). Consequently, the integrity offers a backbone for increasing the sustainability value of an organisation. However, the operationalisation of integrity at different levels of an organisation remains vague, although the management gurus such as Stephen Covey, Peter Drucker and Manfred Kets de Vries treat integrity as a quality of management (Van Maurik 2001).

The definition of integrity in the Merriam Webster's online dictionary is (1) ‘firm adherence to a code of especially moral or artistic values’, e.g. when all the personnel within an organisation are aware of the significance of following up internal and external standards, regulations and procedures in relation to HSE and running the operations; (2) ‘an unimpaired condition: soundness’, e.g. when an organisation is occupied by competent and skilled people, appropriate processes and methods, technology and (3) ‘the quality or state of being complete or undivided: completeness’, e.g. when all the personnel in an organisation are aligned with stated strategies and policies.

The word ‘integrity’ originates from the Latin word ‘integritas’. However, ‘integrity’ not only corresponds to ‘morality/ethics’ (Newman 2003, Lowe et al. 2004) ‘honesty’, ‘justice/respect’, ‘not corrupt’ (Den Hartog and Koopman 2002, Peterson and Seligman 2004), ‘openness/authenticity’ (Peterson and Seligman 2004, Koehn 2005) but also (Solomon 1999, p. 39) to ‘wholeness’, ‘completeness’, ‘purity’ and ‘integratedness’ (see, for example Taylor 1981, Benjamin 1990, Montefiore and Vines 1999), ‘word/action consistency’, (Bews and Rossouw 2002, Worden 2003), ‘absence of unethical behavior’ (Craig and Gustafson 1998, Posner 2001, Mumford et al. 2003). Integrity is particularly essential when an organisation is surrounded by conflicting stakeholder demands: searching for a balanced and optimal performance especially when the organisational values and interests are mutually exclusive.

Companies are obliged to maintain their integrity mainly due to moral, legal and economic reasons as breaches of integrity undermine their competitive advantage. Due to moral reasons, companies are trying to reduce stakeholder and regulatory force by preventing dishonourable conduct and encouraging honourable conduct. Due to legal reasons, companies are obliged by law to assure their own integrity (e.g. design, operational and technical) and that of the workforce. The Statistics of Ethics Resource Center in 1996, illustrates that about 29% of employees feel that they are under pressure because of managers' own ethical standards, which are aligned with company objectives (Kaptein 2000). Due to economic reasons, companies are not supposed to tolerate or admit unethical practices or create a breeding ground for such practices. Breaches of integrity cost money and honourable practices increase goodwill. For example, a multinational company such as Shell lost money and goodwill when violating human rights issues in Nigeria and when trying to sink the Brent Spar O&G platform in the UK sector of the North Sea.

Lanquetin (2005) argues that the aim of IM in relation to floating O&G production units is ‘to ensure management and continuous follow up of floating units from a safety, environment, operational, maintenance and quality management’. The Federal Energy Regulatory Commission (FERC 2005) defined IM requirements as ‘to assess, evaluate, repair and validate, through a comprehensive analysis, the safety, reliability and security of their facilities in high consequence areas to better protect the public and the environment’ (FERC 2005).

Based on the above discussion and views, this article adopts a definition of AM proposed by the BP (2004).

Definition 3 (IM): IM is ‘the application of qualified standards, by competent people, using appropriate processes and procedures throughout the plant life cycle, from design through decommissioning’.

The meaning of integrity at industrial plant level

The concept of integrity is, in practice, mostly confined as a characteristic that only human beings can have (also see Taylor 1981, Becker 1998). However, the concept of integrity is surfacing in industrial practices based on the neighbourhood that the particular industry belongs to. In relation to the O&G industry, for example the concept of integrity has appeared at plant level in terms of AI and it has further cascaded down in to design, technical and operation integrity (De Jong et al. 2009).

In order to reach projected integrity, it is vital to study the way in which the various expectations of the workforce are aligned with one another (see also Ratnayake and Markeset 2010a). This is mainly due to expectations or demands from the world outside an organisation that may be anticipated, but not always coincide with the expectations that have developed within an organisation. As a result, companies regularly appear to be lagging in fulfilling the stakeholders' demands, alleviating the sustainable value to them. In this case, the organisation is said to be lacking in integrity.

However, once an organisation is based on production or manufacturing operations, then the integrity of plant level operations reflects the integrity of the whole organisation and vice versa. Figure 1 illustrates the closed-loop iterative process of the manner in which the stakeholder interests' influences are transmitted through the company to the operations and maintenance of the plant assets (see also Ratnayake and Liyanage 2007). The figure also illustrates how stakeholders influence governing documents, plant management and the organisation. The thickness of the lines and directions of the arrows indicate the strength and direction of the influence (thicker lines, stronger influence). Thus, in manufacturing or production operations centred businesses, the integrity of plant level assets decides the overall performance of the business organisation.

Figure 1 The influence on performance of industrial plant from internal and external stakeholder demands (also see Ratnayake and Liyanage 2007).

In this context, the industrial plant is treated as a group of PA that is interacting with other assets (e.g. financial, information, intangible). Furthermore, the HA are a major part of the performance of the various cross-discipline processes and activities involved in operating, maintaining, modifying, etc. the assets. An attempt at illustrating this notion is shown in Figure 2.

Figure 2 Notion of AI (adopted from Ratnayake and Liyanage 2009).

The outcome of good AIM supported by effective management systems includes meeting defined performance standards for effective control of various financial, societal and environmental threats. Rondeau (1999) has, for example defined AI in relation to infrastructure resource planning and investment delivery objectives as the ‘sum of all those activities that result in appropriate infrastructure for the cost-efficient delivery of service’. This definition relates to activities involved in the management of the assets. Similarly, Pirie and Ostby (2007) holistically relate AI to

a continuous process of knowledge and experience applied throughout the life cycle to manage the risk of failures and events in design, construction and during operation of facilities to ensure optimal production without compromising safety, health and environmental requirements.

Richardson (2007) defined AI in relation to maintenance discipline as ‘the maintenance of fitness for purposes of offshore structures, process plant, connected pipelines and risers, wells and wellheads, drilling and well intervention facilities and safety systems’.

In this article, we choose to adopt the AI definition from the KP3 Project Handbook (HSE UK 2007):

Definition 4 (AI): ‘Asset integrity is the ability of the asset to perform its required function effectively and efficiently whilst safeguarding life and the environment’.

For instance, the integrity of an offshore O&G production facility can deteriorate over time or result in substandard performance due to wear and tear, accidental damage, extreme weather conditions, geotechnical/geological hazards, modifications in technology or use, etc. As stated in HSE UK (2007), some of the key issues that are significant in managing the integrity of O&G assets include to

•	consider HSE risks of all assets at the design stage, operational level and technical condition control stage;
•	analyse issues that can be harmful to the organisation;
•	take proactive measures to prevent and control undesired events;
•	evaluate whether personnel are trained and competent enough and
•	monitor, inspect and audit to verify that the desired levels of performance are achieved.

The sub-categories of overall AI include design integrity, technical integrity and operational integrity. In this article, we adopt the definitions by De Jong et al. (2009) as they closely resemble our view of design, technical and operational integrity.

Definition 5 (Design integrity): ‘assurance that facilities are designed in accordance with governing standards and meet specified operating requirements’.

Definition 6 (Technical integrity): ‘appropriate work processes for inspection and maintenance systems and data management to keep the operations available’.

Definition 7 (Operational integrity): ‘appropriate knowledge, experience, manning, competence and decision-making data to operate the plant as intended throughout its life cycle’.

The management of these three categories of AI should lead to safe processes and securing the overall AI of an asset-centric organisation (De Jong et al. 2009). Figure 3 illustrates the notion of AI in relation to an asset-centric organisation.

Figure 3 Physical industrial plant assets in relation to other critical assets in an asset-centric organisation.

AI may decline due to, for example ageing of equipment, loss of knowledge through retirement of experienced and skilled staff, lack of knowledge and competence, lack of awareness about frequently changing procedures and standards, lack of transmission of sustainability concerns at the plant-level operations, etc. Hence, management of AI is necessary throughout the life of the asset from the initial design phase, through manufacturing, installation and commissioning phases, operations phase as well as in the retirement and the recycling phase.

Asset integrity management (AIM)

The shape of the developed and developing expectations of an organisation depends on the nature of the routine employees' performance. This is because human intelligence cannot be replaced with technologies or technological sub-systems (Koch 2002). In a formal organisation, the expectations are explicit and formalised in rules and procedures while they can also develop in informal way.

For example, the AI of an asset-centric organisation depends on the ability of the management to set the right direction, where the unwritten or written organisational expectations may or may not be well understood by the employees of such an organisation (Drazin and Sandelands 1992). Hence, the integrity of organisations' assets is a reflection of the awareness, encouragement, etc. of the personnel to operate them in a responsible and knowledgeable way to satisfy stakeholder demands. However, the integrity of the organisation's assets is the result, which is not the total or the average of the integrity of the personnel (Kaptein 2000).

The necessity of AIM arises when the internal expectations are not aligned with stakeholder expectations, resulting in a harmful atmosphere to the company or its operational environment leading to breaches of integrity. The violations, or breaches, of integrity within an organisation can happen within the different levels in the organisational hierarchy. For instance, some of them include abuse of power, discrimination, leaking confidential information to the press or competitors, the reckless use of the organisation's assets, etc. Besides this, integrity violations may happen due to knowingly selling defective products, hiding financial failures, extorting from suppliers, spying on competitors, evading environmental legislation, etc.

Another facet of integrity violations is due to shortcomings in personnel guidance. Although blame may be assigned to the company in question, blame cannot be tackled solely at the level of the individual employee. Therefore, in developing the integrity of an organisation, it is paramount to measure how the various expectations that exist in and around the company are in line with each other. The breaches or violations of integrity cannot be entirely prevented. However, by taking measures at the organisation level, the management can ensure that the damage to the company and its environment is limited.

In this context, AIM is mainly concerned with gearing up stakeholder and organisational expectations by bridging the gap between them. In an asset-centric organisation, this is accomplished at the plant level, by means of improving three types of relationships from the point of view of AI, namely

1.	The relationship between the organisation and its stakeholders (i.e. when the interests and expectations of the stakeholders (financial and HSE) are incompatible with the interests of the company) through the refinement of plant strategies, policies, etc. [strategic/top management].
2.	The relationship between the workforce and the organisation (i.e. when there are conflicts between the functional interests of employees, managers, departments and units) by bridging the awareness gap between the workforce's understanding and organisational expectations [tactical/middle management].
3.	The functional relationship between workforce and activities (i.e. when there is a discrepancy between the personal interests of the employees and the things that should happen at the operational or the activity level of the company) [functional/process/operational management].

All three sub-categories of AI are analysed in relation to levels (1)–(3). Figure 4 illustrates the notion of translating stakeholder demands to the plant level operations via operational, technical and design integrity.

Figure 4 Translating stakeholder demands to plant level operations (see also Ratnayake and Liyanage 2007).

Having this point of view for managing the AI at plant level, the authors adopt in this article the definition of the AIM proposed in the KP3 Project Handbook (HSE UK 2007).

Definition 8 (AIM): ‘Asset integrity management is the means of ensuring that the people, systems processes and resources which deliver the integrity, are in place, in use and fit for purpose over the whole life cycle of the asset. Whole life cycle comprises: design, construction, installation, commissioning, and operation’.

AIM: improving asset performance

When the management has an insight into the organisational causes (i.e. kind, nature, extent) of the integrity violations, it is easier to take effective measures for mitigating them. The measurement of integrity performance can assist in clarifying and unravelling the implicit and explicit, internal and external (sometimes conflicting) expectations confronting the personnel. The personnel guidance can be improved through further trainings, workshops, etc. based on the measured and analysed expectations of the potential workforce.

AIM tries to improve the performance of PA through increasing the awareness of the workforce to realise the legitimate fundamental expectations of stakeholders. Merely having a written code of conduct may not be an appropriate instrument to achieve this due to the frequent changes of stakeholder demands. However, among other things, AIM should focus on the creation of conditions within which an organisation-wide consciousness-raising effort and internal interaction can take place. This will reduce the variability and consequently improve the assets' performance. Figure 5 illustrates the notion behind the improvement of the asset performance through increased awareness not only about stakeholder demands but also about system parameters and behaviours.

Figure 5 The notion of improving asset performance with increased awareness and reduced system variability.

Having reduced variability, the collective insights that can arise from organisation-wide discussions should focus on understanding one another's problems within the organisation and getting insight into the different opinions that employees have. This will enable the development of specific activities to find the right balance between the conflicting interests and expectations that confront organisations and their stakeholders. By examining the organisation from this perspective, it is possible to work on improving the organisation's AI.

Figure 6 illustrates the possible balance that should be achieved on asset performance through ‘triple bottom line’ lenses. For instance, asset performance could be related to (1) finance (equipment failure may lead to the asset not being capable of producing a saleable product), (2) environment (equipment failure may lead to a breach of an environmental standard, regulation or license condition) and (3) health and safety (equipment failure may lead to the injury or death of employees or other people).

Figure 6 AIM: balancing conflicting stakeholder (triple bottom line) demands.

As both economic and non-economic factors are involved, the decisions regarding assets' performance should be seen through balanced and multiple criteria (see also Kaplan and Norton 1992, Blanchard and Fabrycky 1998, European Foundation for Quality Management 2003, Al-Najjar and Kans 2006, Ratnayake 2010). There are several frameworks, which view business performance through more than one perspective to mitigate the weaknesses of traditional, uni-dimensional and backward-looking nature measurement systems (see Andersson et al. 1989; Eccles 1991, Lynch and Cross 1991, Kaplan and Norton 1992, 2004, 2006). However, these frameworks do not provide a formal mechanism to analyse the performance. A framework and a mechanism to address this issue are presented later.

An approach and a framework for measuring AI performance

AI refers to the degree to which assets satisfy the legitimate expectations of the surrounding world. However, it is not possible to fulfil all the expectations. Those expectations that are widely supported and generally regarded as appropriate and essential can be regarded as legitimate expectations or key strategic measures. ‘The balanced scorecard (BSC) is the most widely used performance management system today’ (Johnson 2011). ‘It intensifies the focus on the strategy and identifies the management and organizational actions required to get performance back on track’ (Kaplan and Norton 2000a). Moreover, the BSC identifies five principles for strategy management in the book The Strategy-Focused Organization: How Balanced Scorecard Companies Thrive in the New Business Environment: ‘mobilize, translate, align, motivate, and govern’ (Kaplan and Norton 2000b).

The book published by Kaplan and Norton (2008), The Execution Premium: Linking Strategy to Operations for Competitive Advantage, discusses the former principles in depth. Moreover, Strategy Maps: Converting Intangible Assets into Tangible Outcomes focused on the second principle ‘translate’, in which it described and illustrated how strategy maps and scorecards could be customised to many different strategies (Kaplan and Norton 2004). The second principle, ‘align’, i.e. how to create and capture corporate synergies through vertical and horizontal alignment of business and support units, is discussed in the book Alignment: Using the Balanced Scorecard to Create Corporate Synergies (see Kaplan and Norton 2006, Ratnayake and Markeset 2010a). The same book also contained material on the fourth principle – ‘motivate’ (and align) employees for strategy execution in their business or support units. Double-Loop Management: Making Strategy a Continuous Process provides a discussion about the fifth principle – ‘govern’ – to make strategy a continual process (Kaplan and Norton 2000a).

However, the first challenge in AIM strategy execution is to be sure that all of the ‘sub-goals’ and ‘departmental action plans’ are themselves aligned with the larger strategy (Ratnayake and Markeset 2010a). Kaplan and Norton (2004) discuss this challenge in the strategy mapping process. In that process, a hierarchical list of strategy drivers is mapped onto the company strategy. At that point, it is relatively straightforward to develop targets and action plans for each driver. The mapping process, however, is not as formalised as the analytic hierarchy process (AHP) methodology (Saaty 1980, 1990). Also, the attention paid to integrating enterprise risk management as well as the evaluation of priorities in relation to the organisational strategy is not sufficient in the strategy maps or BSC approach, while it is in AHP. Hence, the AHP approach is a potential candidate for AI-related performance measurement and/or evaluation. Furthermore, this approach has the ability to synthesise data, experiences, insights and intuitions in a logical and thorough way for making the optimum decision (Saaty 1980).

Moreover, the integrity is a relative concept, and acceptable assets' performance today may be considered unacceptable tomorrow. Therefore, it is vital to have a dynamic mechanism to integrate industrial data and experts' experiences, intuitions and intentions in a logical and thorough way based on influence factors and circumstances. The mechanism behind AHP provides an opportunity for integrating the aforementioned challenges. Hence, in order to measure AI performance in relation to sustainability demands, the framework shown in Figure 7 is suggested.

Figure 7 The framework for measuring AI performance.

The sole idea of using AHP-hierarchical structure is to incorporate a balanced set of measures (e.g. financial, health, safety, environmental) within the measurement scheme and to incorporate industrial data and experts' experiences, intuitions and intentions in a logical and thorough way. Based on the developed hierarchical structure as suggested in Figure 7, an interview questionnaire can be made (see Ratnayake and Markeset 2010b) in order to assess the experts' performance.

The framework advocates an approach to self-sustained success in AIM. It is achieved when the top-down managerial direction, priorities and performance goals are clearly aligned with the bottom-up delivery capabilities and middle aligned with behaviours and organisational factors. The assets' integrity is cascaded down to a manageable level in terms of DI, OI and TI. In the suggested approach, the performance measurement is conducted on the basis of interviews with key experts supported by written documentation (see Ratnayake and Markeset 2010a). As proposed by Saaty (1980, 1990), the first and most important thing in the AHP is to develop a hierarchical structure based on the dilemma under consideration. For instance, Ratnayake (2010) as well as Ratnayake and Liyanage (2009), illustrate a case study carried out for developing a hierarchical structure for OI management.

Apart from that, Ratnayake and Markeset (2010b, 2010c) discuss how to develop a hierarchical structure in relation to triple bottom line and strategic dimensions, respectively, to manage the TI of oil and gas production and process facilities located on the NCS. Moreover, Ratnayake and Markeset (2010a) provide comprehensive analysis about developing a hierarchical structure and its implementation using a case study carried out in a leading automobile manufacturing plant (located in the USA) to measure organisational alignment with respect to newly implemented company policies.

In the case of a multinational company, assets are located at geographically different locations. Hence, one industrial plant may be performing differently as compared with another. Thus, an effective hierarchical diagram can only be developed if the management has an insight into not only the ‘type’ and ‘extent’ but also the ‘nature’ (i.e. the organisational causes) of breaches of integrity. Measuring the integrity performance can help clarify and unravel the implicit and explicit expectations as well as the internal and external (conflicting) expectations confronting employees. The measured and analysed results (i.e. through prioritisation and gap analysis) can form the basis for taking concrete steps to improve the workforce guidance. This enables continuous improvement and sustainable asset operations.

Four different dilemmas for developing the hierarchical structures for critical investigation of AI have been identified as shown in Figure 8 (i.e. design integrity (DI), technical integrity (TI) and operational integrity (OI) composed of and supported by HA, PA, FA, IA and ITAITA).

Figure 8 Four integrity-related dilemmas.

•	Formal expectations: explicit tasks, guidelines, rules, procedures, etc.
•	Informal expectations: actual norms and values, customs and practices, creditable behaviours, etc.
•	The stakeholder expectations: existing, realised and unrealised expectations such as following up HSE regulations, implementing best available technology to reduce the burden on society and the environment and implementing zero discharge concepts.
•	Conflicting expectations: dilemmas caused by the incompatible expectations such as increasing profit margin minimising HSE burden.

Discussion

The report about AI, published by the offshore division of health and safety executive's hazardous installations directorate (see KP3 2007), reveals that the influence of the engineering function had declined to a worrying level as a result of technical authorities being under pressure. Consequently, the technical authorities are often reacting to immediate operational problems rather than taking a strategic view to provide expertise and judgement on key operational engineering issues (KP3 2007). Hence, it is vital to manage change. In order to manage the change, it is paramount to measure experts' performance in order to assure AI, as a single human error can destroy all the sophisticated technology, leading to uncontrollable circumstances.

For instance, in the event of the Macondo well blowout in the Gulf of Mexico, although there were several safety barriers to prevent a blowout with the help of very sophisticated technology built-in, lack of change management, limited awareness of the risk and impact of blowouts, attitude, trigger regulatory changes, etc. (see Rygg 2010) led to the oil and gas blowout, leading at once to a fire and explosion, destroying all the sophisticated technology. This led to the loss of 11 workers' lives as well as the leak of millions of barrels of crude oil to the environment, resulting in more or less unrecoverable environmental damage as well as health and safety damage to the surrounding society while disturbing their day-to-day living activities.

In addition, the UK government had to support BP in order to safeguard pensioners in the UK as most of them had ‘pension funds who rely on their investment in BP as part of their financial security and in many cases their retirement income’ (see Upstreamonline 2010). The event of BP becoming bankrupt would have had an effect on UK society. Furthermore, if BP had been unable to pay for the environmental damage in terms of recruiting fishermen to the company for different job tasks (e.g. cleaning jobs) and different processes to clean the natural environment, then it would have affected the fishing and surrounding community in the USA.

Conclusion

Using the framework suggested in this article, it is possible to measure asset performance in relation to experts' awareness, in which an organisation's management can achieve a right balance between the conflicting interests and expectations that confront the industrial workforce. The focus in this article is to establish tangible and achievable gap analysis in terms of organisational issues in relation to assets' integrity assessment. By examining an organisation in this perspective, it is possible to improve the AI of the organisation. This is the prime objective of AIM and continues improvement (or sustainability). Basically, if you can measure AI, then you can manage it. However, realising all the legitimate expectations of stakeholders will often not be possible due the causes of problems of the top/strategic middle/tactical and functional/low level management.

The extent, depth and method of AI performance measurement depend on conflicting dilemmas due to stakeholder demands and the wishes of the industrial organisation concerned. However, assessing the possibility of carrying out measurement in relation to other components, subsystems or systems separately reveals that various component combinations may provide additional information. Such a holistic measurement scheme can provide information about

•	places in the industrial organisation where there is a risk of conduct that lacks AI, owing to lack of awareness of adequate procedures and rules, etc.
•	the effectiveness of existing procedures and rules;
•	the gaps in relation to formal organisation that are not covered by the informal organisation;
•	significant causes of conduct that lacks integrity;
•	vulnerable tasks, activities and departments;
•	the AI dilemmas that confront employees;
•	stakeholders with grounds for complaint because the organisation violates their legitimate interests and expectations;
•	a discrepancy between the expectations of stakeholders and the efforts of the organisation and
•	the measures to be taken and activities to be developed to improve the organisation's integrity.

By examining an industrial plant in this perspective, it is possible to work on improving the plant's integrity and to achieve sustainable industrial operations. Consequently, it will continuously improve the integrity of the organisation, which is centred on plant's (or plants') performance. Furthermore, the approach will enable the translation of the stakeholder and organisational expectations to plant-level operations and vice versa.

This article is based on the ideology that human intelligence cannot be replaced by sophisticated technology. However, faulty sensor data in the context of real-time condition monitoring of dynamic assets (e.g. pumps, turbines) as well as wrong interpretation of degradation data in the context of static assets (e.g. piping components, vessels) generate significant challenges to manage the AI and achieve sustainable operations on ageing assets (e.g. production and process facilities). Hence, further research should be carried out to analyse the reliability of the aforementioned data available for human (or experts') decision making.