Avoiding the 'Swiss cheese' effect

April 2, 2011

Cheese may never be far from the thoughts of the French. But what concerns classification and verification group Bureau Veritas is the holes in cheese – and more precisely how to avoid them. OE hears from BV offshore projects manager Jose Esteve about the company's latest thinking on asset integrity management, especially for today's ageing FPSO population.

Asset integrity management is about avoiding the Swiss cheese effect,' explains Jose Esteve, offshore projects manager at Bureau Veritas.

‘You have to stop the hazards finding a way through the various mitigation and prevention mechanisms you put up. What we have learnt with our oil major partners is that successful asset integrity management is much more than a set of tools and much more than a set of inspection regimes. It is a philosophy. You need the tools, you need the infrastructure, you need a clear definition of responsibilities. But above all you have to visualise and understand the problem fully, which tools alone cannot do. And we have also learnt another important lesson, and that is that the later you start, the harder it is to do properly.'

Bureau Veritas is involved in one way or another with the provision of integrity assessment and integrity management for around 100 major offshore units worldwide. Most of them were brought into Bureau Veritas integrity management once already in service. But a dozen of these units were covered from the outset by BV's VeriSTAR AIMS tool.

This is a web-accessible database with a hull structure viewer married to an inspection system and a set of analytical tools. ‘Operators can choose whether they use the database only as a handy means of storing all the information on the unit, or they can go further and design the inspection plan to fit with it,' says Esteve. ‘That brings all the disciplines of reliability engineering to bear to guide the choices of what is inspected, how it is inspected and how often it is inspected, and how often preventative maintenance is done.

‘We realized that when a project team passes a newbuild to the operations people, a lot of useful information for integrity management is not properly handed over,' Esteve explains. ‘One such item is an adequate inspection plan. It has taken several projects to get the actual inspection and maintenance guys involved in the loop during the creation of the hull and topsides inspection programme. After we did the risk-based inspection (RBI) plans for the topside structure of the Dalia FPSO we realized that the final users of it were providing operational feedback that would have optimized its preparation. The same thing happened with the hull RBI program done for the Unity FSO. Their input is necessary if you really want to make sure the final deliverable will be used. It is also a very good way to create and establish an adequate integrity management state of mind.'

Key steps
Implementing an asset integrity management approach is particularly difficult for the hull or the topside process structure, Esteve points out. BV has chosen to follow the SIM (Structure Integrity Management) process described in the international standards for offshore units (ISO, API). The process has four key steps: data management, assessment and evaluation, strategy design and implementation.

Data management is a key element of any AIMS system and it involves not only design and construction data but the natural feedback from the life cycle. How will you build in a learning process so that continuous improvement is part of the system? It is at the assessment and evaluation stage that a decision needs to be made.

‘Of course you need good tools to make this work,' says Esteve. ‘For example, we developed a SIM-based methodology for the RBI plan of jacket structures and topside structures. We soon reached the conclusion that in order to jump from theory to practical use we also needed to build the adequate software tool. A tool capable of working as data storage but with intuitive 3D representation and multi-level display including risk levels, inspection data entries, scantlings and structure break-down. Bureau Veritas is currently using these tools for the provision of a structure integrity management programme for the Bayu Undan platforms in the Timor Sea.

‘But tools alone won't help you if you don't get the thinking right and follow the steps properly. You need to be clear about your objectives, and clear about the chosen route. Will you go for a prescriptive system, or a goal-based system? RBI programmes always seem attractive but they are only as good as the accuracy and completeness of the input data.'

Based on the selected path the actual strategy is drafted followed by the actual operational implementation on the unit. The integrity manager has to set the availability and safety criteria for the asset, and risk ranking and prioritising flow from that. The class, state, verification and company requirements have to be married up with an inspection scheme and a system which is accessible to all the partners and departments involved.

‘That sounds obvious,' notes Esteve. ‘But it is remarkably hard to achieve in practice for a major project, which may involve several different departments in the operating company alone, such as the project team, the safety department, marine operations, maintenance is someone else's job, and all these need to liaise with several external parties and regulators. That means close attention to data management, and data input and accessibility. And it calls for choices – do you store raw inspection data, or processed reports, for example? Do you bring together your company planned maintenance scheme with a RBI schedule? If so, how do you do that? The objective of AIMS is always the same, to reduce downtime and cut out unexpected incidents and unplanned stoppages, but there are many routes to that goal.'

Esteve says that there seems to be a trend to ask for very long design lives of 40 years or more and for these units it will be vital to have, from the outset, the right approach and mindset to cope with the natural corrosion of steel in marine environments.

All-in-one philosophy
A similar state of mind is needed for LNG floaters now in prospect, considering their requirements concerning storage, processing and offloading capabilities. ‘But as yet not everyone adopts this all-in-one philosophy, and the consequences can be that problems with plant affect the structure in unforeseen ways, or vice versa. So start early, plan everything in, make your choices on priorities, then select the right tools. Remember that you get what you inspect, not what you expect,' he says. ‘You have to identify the initiators or triggers of small failures which can lead to bigger problems, and ensure they are looked for and monitored to catch them at early stages if no other preventive action is available.'

Part of an AIM state of mind is the importance of RAM and FMECA. RAM stands for Reliability, Availability, Maintainability – which sound important, and are, says Esteve. But there is a danger at this stage that the model and the process obscure the vital reality. This process is based on the data that goes in, on the equipment, its use, its history and its previous failures. If that data is flawed, then the whole process going forward is weakened, because the wrong decisions on inspection programmes and conditionbased maintenance will be made.

BV uses a Monte Carlo simulator program, Optimise, to simulate and assess the risk and model lifecycle availability. It has been used in over 100 major projects so far. ‘It is important to take time at this stage to get comprehensive and correct data to underpin all the later decisions,' says Esteve. ‘Check that the failure data is based on exactly the same technology, the same type of service, the same equipment and a period of stable service. All this may be tricky for new LNG units, but in fact much of the equipment in use is well documented from shore applications and gas carriers, so the only new variable is its combination with the marine environment on floating units. And that is the tricky part.'

FMECA (Failure Modes, Effects and Criticality Analysis) will establish the consequences and seriousness of any individual plant or structure failure, which can be quantified with the probability and frequency of occurrence to determine the risk level. According to Esteve, all the data analysis and modelling leads to a very detailed reliability-centred maintenance plan for the plant and equipment and a risk-based inspection plan for the unit structure intended to detect early signs before a small crack becomes a major problem.

Defining how inspection should be done and how it should be recorded and analysed is important, as standardisation means multiple parties can all understand the possible consequences of any defect in the same way. ‘That is where tools help,' says Esteve.

‘We have a system to turn thickness measurement into a graphical interface, for example. The graphical visualisation tool gives AIMS users a 3D view of the unit's structure, with hot spot mapping, access to all thickness reports, inspection reports and photos of each part of the structure. Users can walk through the unit on screen to see what has been done and what needs to be done and where. The level of detail on view is tailored to what the operator wants and will depend on who populates the database. Those major units covered through the newbuilding stage should have got a full history from the building, through to in-service inspections.

‘It is easier to use and on-site managers, regional managers and head office can all see the same thing and discuss things without paper travelling around the globe.'

Asset integrity management is a living concept and requires constant reviewing and feedback, adds Esteve. ‘Experience shows that orphan items of equipment or orphan processes are the ones that slip through the holes in the Swiss cheese and cause the incident or downtime the whole system is in place to avoid. So my one big message to all integrity managers is to use good tools, but don't be dazzled by them. Take a good hard look at the system and check the tools are getting the right information and data. And ensure that your team has a state of mind in line with yours.' OE

Current News

Norway Delays 26th Oil Licensing Round

Norway Delays 26th Oil Licensing Round

Exail's DriX USV Gets Lloyd's Register Certification

Exail's DriX USV Gets Lloyd's Register Certification

UTEC Showcases New Unmanned Catamaran for Offshore Wind Surveys

UTEC Showcases New Unmanned Catamaran for Offshore Wind Surveys

Windcat Orders Two Hydrogen-powered CSOVs from Damen Shipyards

Windcat Orders Two Hydrogen-powered CSOVs from Damen Shipyards

Subscribe for OE Digital E‑News

Offshore Engineer Magazine