Article-V16
Download as DOCX, PDF1 like146 views
1) The document discusses developing an effective asset management information strategy and system using quality data and standards. 2) It identifies common flaws like lack of understanding knowledge utilization and issues implementing information systems or with data quality. 3) The proposed solution is an Asset Integrity Management information framework that standardizes data using ISO standards and structures it in a way that prevents bad decisions from low quality data.
Article-V16
- 1. One implicationof this increasinglydemand- ing market is the necessity to develop a strategy of speedy adaptation and accurate decision-making. This includes an infor- mation strategythat results in the creationof knowledge incorporating all available data, information and experience, however, a con- tradiction here, in that the process of knowledge creation is typically misused and misunderstood. The source of this problem lies in the knowledge creation process, where three main areas for AM have been identified as common flaws : Lack of corporate understanding and knowledge utilization The inability to implement and utilize an asset management informationsystemas a solution. Data quality problems associated with asset management, analytical and prognostic models. Corporate understanding and knowledge utilization Despite all the ongoing discussion concern- ing importance of human and intellectual capital, few managers understand the true nature of the knowledge-creation process. This is because they misunderstand what knowledge actually means andwhat compa- nies must do to ensure that it remains a valuable resource for increasing competitive advantage. Where the sharing ofknowledge is not considered tobe inanybody’s interest, challengingdiscussions are oftenseen as at- tacks uponindividual competence or ability. The role of the manager in eliminating this problemis ofteneither neglected or not rec- ognized at all. An organization needs to persuade its staffto share, combine and dis- seminate knowledge and information. Only when this new or improved knowledge is seen in the right perspective and managed by an effective information system will it have the potential to become a critical re- source for effective decision-making. The model represented in Figure 1 is based on Nonaka’s SECI model, the primarypurpose of which is to gain understanding of those activities that transform data into useful practices. Failures in the implementation of AM information systems. Developing an AM Information strategy to suit the business needs of customers is not merelyabout finding technology, whether by using state-of-the-art IT-based solutions or implementing the best-in-class software, but rather in making the whole thing a strategic consideration. Despite significant investments, stakeholders, like executive management and users, are very often not involved enough, giving rise to difficult issues, such as “silo mentality” being completely ignored. Projects failto be deliveredontime, on budget or with their intendedfunctionality because ofhavingtoowide scope, resultingin an unmanageable complexity, incompetent resources and no clear business objectives. Data Quality Issues for analytical and prognostic models To understand the context, characteristics and complexity of AM prognostic models, it is necessary to develop a more detailed understanding of the underlying failure processes in human and asset behaviour, along with other relevant factors. Once the structure of a model has been determined, the extent to which data, information and knowledge influence the outcome of the model should be verified by actual observations. The purpose of developing an advancedAM analytical or prognostic model is to estimate the probability that similar assets demonstrate specific performance in the same situation in the future. This estimate, and the understanding of the relationship between its elements, helps in the making ofmeaningful decisions based on quality data that transcend guessing. The accuracy of these models for AIM depends decidedly upon data quality and the extent that an information system provides the relevant data of those elements likely to affect, justify and improve a model. Consolidating Informatization Creation Utilization Optimization © PaulStam Educating Practices Implementing Data Observing Coding Knowledge Information Expert Utilization Analyzing Combining knowledge Sharing Knowledge Knowledge Figure 1 The principle of knowledge creation Quality information, a critical resource for asset management Without high quality information managers can neither understand nor adequately respond to real world events. Making the right decisions in Asset Management (AM) will invariably improve the value in the lifecycle of an asset, but is far less likely if reliable information is unavailable. If data does not fully relate to the real world, then they are useless for analytic and prognostic modeling within Asset Integrity Management (AIM). By specifying the four crucial buildingblocks for an information system stand- ardized by the use of the ISO standards like the ISO 55000 and placingit in the organizational context, utilizingdata for information becomes a realistic option for asset management. The principle of knowledge creation Consolidating: the process of data acquisition and the storage of observations (i.e. measurements , inspections and audits) and catego- rised knowledge. Utilization: the process of consol- idating information and putting the model, prototype or operation-al mechanism into practise Utilization: the process of consol- idating information and putting the model, prototype or operational mechanism into practise. Informatisation: the process that involves data analyses, the uses of models and expert knowledge in or- der to transform data into information. Creation: the continuous process of discussing, challenging and com- bining all available information into tacit knowledge based concepts (i.e. model, prototype and operational mechanism).
- 2. The use of standards The use of standards not onlyprovides a ba- sis for business collaboration through an exchangeable and reliable standard ofqual- ity, but canalso be seenas piece of industry- specific knowledge capital. To exploit this advantage in the development of an information system, four potential stand- ards have been identified, i.e.: ISO-55000 for directing the AM infor- mation requirements. ISO-14224 for providing the reliability of data requirements. ISO-13381 to gain insight into the data re- quirements for prognostic modelling. DuPont modelto structure the asset value data. The four buildings blocks for quality information. The AMinformation framework(Figure 2) is designed to be the first steptoward building an information system that provides quality data to the organizational context and also meets the requirements of stakeholders. The fact that data qualityis not onlylimited to IT related requirements but also ad- dresses the quality requirements of AM relevant ISO standards and modelsis unique in this context. Making a wrong decision based on bad data will be less likely when experts can recognize the flaw. However, if common sense does not protect us from making flawed decisions in critical situa- tions, then the alternative option is to manage data that provides informationthat does not contradict the real world. Inother words, the use of an AIM information framework hasthe potentialto prevent the “garbage in gospel out” problem, and pro- vides the abilityto make information-based decisions. It is also very much the second- best option. Asset Management is not the exploitation of quick wins Although the use of the framework sounds appealing, it comes at a price. The complex- ityinherent in using modelsto estimate the behaviour of equipment not only requires highlycompetent staff, it alsoneeds a valu- able information system over a longer period for managing and utilizing the data. Assuming that we have todeal with the sto- chastic or random uncertaintybehaviour of equipment, this periodoftime is contingent on the failure rate of components and on the necessary accuracy of the information. In general these periods measured in dec- ades rather than years. The process from concept to model and finally a common- database for AM. To translate the abstract concepts of the informationframework intoa commondata reference model, four techniques are applicable: The first is data normalization to struc- ture and sustain data integrity. The second is derived from Quality Functional Deployment (QFD). In contrast to QFD, its purpose is not to focus on the most important client demands, but to define the role (as an actor) in the relations between requirements of ISO 55000 and data classes (see additional information). The outcome is the source for defining entities inthe data reference model. The third technique is the semantic analysis for defining the relations be- tween attributes (data fields from tables). The last technique is the multiple rela- tionship analyses used for solving the integrityviolation in case ofmultiplere- lationships between unique key attributes. Conclusions There is little doubt that information and knowledge is a strategic resource, although it remains doubtful that the process of knowledge creation is sufficiently well understood. There can be no doubt, how- ever, that an information system is the necessary insurance against GIGO occur- rences. The proposed solution embodied by the AIMinformationframework has the poten- tial to make real progress inthe creationof strategic, relevant information for AM. GIGO ‘Garbage in, Gospel out’ is a more recent ex- pansion of the acronym. It is a cutting comment on the tendency to put excessive trust in "computerized" data, and on the tendency for individuals to blindly ac- cept what the computer says. Figure 2 AIM Information Framework 1 The strategic asset management plan (SAMP) is the foundation for information re- quirements, which are linked to clear business objectives. 2 The AIM information strategy de- scribes how to manage, maintain and improve strategic relevance according to the SAMP. The scope of this approach is defined by Integrity (reliabilty) requirements with re- spect to the context of the organization and business environment. 3 The AIM information system itself can be seen as the processing layer where soft- ware, hardware and human activities together transform quality data into relevant information. 4 The purpose of the AI dataset is to rep- resent the real world in a standardised structure in order to comply with the data re- quirements of the AIM information system. The 4 layers of the AIM information framework Data classes. The process sequence for the mi- gration and integration of data is directed by the origin and interdependency of five data classes. Metadata for data mapping (i.e. table, field, field specification) Object data for the specification of an organization, asset, system, equipment, component or persons. Activity data, or data that describes the activities in assets such as designing, planning, purchasing, con- structing, changing, repairing, operating, monitoring and inspecting. Observation data are the measurements, values or reviews related to an object and the associated cir- cumstances during the acquisition process (e.g. time-frame, state of an object, temperature, flow) Data Quality Directives are the rules and instructions that guarantee data quality (e.g. accuracy, confi- dence and completeness) according to its purpose.