Digital Energy Journal- How WorleyParsons is managing the data lifecycle

Join | Existing Member Login

You are Home » News » View Article

How WorleyParsons is managing the data lifecycle

Thursday, April 24, 2014

Dave Kent, Lead Data Coordinator WorleyParsons London, is developing new ways to manage the data lifecycle in large engineering projects, so the same data is used for the lifetime of the project

Engineering company WorleyParsons is aiming to develop better data management systems for its projects, so the same data can be used through the lifetime of the project.

In the old model of working, the front end engineering was 'almost like a silo of work,' said Dave Kent, Lead Data Coordinator WorleyParsons London (currently working on a North Sea Platform Project).

He was speaking at the Digital Energy Journal Aberdeen conference on November 27, 'doing more with offshore engineering data'.

'Traditionally on a non-integrated document/CAD centric project what would happen at the end is operations and maintenance would get a giant pile of information, and they'd have to scrape [data from the documents],' he said. 'The phases were quite segregated in this setup.'

'Typically you might get 30 per cent of your critical data.'

The modern approach is to gradually collect data and developing an information flow as you go along

With this approach, the operations and maintenance people can end up with over 75 per cent of the critical data in a format they can work with, Mr Kent said.

'If your systems are all integrated you have single point data entry, you have the one piece of data in one place that is shared among all the people that need it,' he said. 'You can avoid having two bits of information for the same item that say something different.'

The data lifecycle for a project is well understood, starting with feasibility studies, moving to detailed procurement and construction, through to decommissioning. On the way, data is also linked to other data, enriched and quality checked. 'It may go through a number of iterations before you have the final data,' he said.

'The best way to approach this is to start off with a bird's eye view.' You need to get an idea of what you are tackling and get some basic processes in place,' he said.

The next step is to define what risks are associated with collecting the data, or not collecting the data, in terms of cost and schedule.

'We need to evaluate the scope very early on. The later you evaluate the scope the more money it can cost you,' he said.

Classes and tags

The most important factors in the data centric world are the class library (the overall structures of the information system), and the tags register (keeping track of all the items and the information relating to them).

With these two in place, you can build software tools which work on the core data, including for drawing piping and instrumentation diagrams, procurements and materials tracking.

You can build intelligent pipeline and instrumentation diagram implementations, procurement, materials tracking. You can make reports and drawings from the data.

Mr Kent describes a 'class library' as an indexing system for information, which can have sub-classes, so for example a parent class can be 'vehicle', sub classes can be 'car, bicycle'.

Each class can have attributes which can go into it, for example a car has an engine, fuel and wheels.

Since a bicycle also has wheels, you might decide to create a subclass of 'wheeled vehicles' containing both cars and bicycles.

'That's a reasonably simple description of what a class is,' he said.

In a similar way you can make a class for 'pump' and have sub-categories below that.

You can re-use a class library from one project to another if they are similar. 'It is important to get the correct class library for the type of project you are doing.

You look at the type of classes you are expecting and the type of attributes you consider to be important. 'By setting up your class library correctly you avoid rework later on,' he said. 'You don't want to get halfway through the project and realise you are missing some key attributes, or you have attributes that no-one is interested in.'

'If you know precisely what you need to collect, this will avoid increased cost.'

The tag numbers are a unique identifier for an item, like a vehicle registration number.

Items can be tagged so you can add related information to the database, for example the spares and servicing needs, connectivity, equipment to operate it, safety requirements, what cables and pipelines it needs.

After the preliminary phases of a project when the layout of the plant has settled down WorleyParsons typically tags locations for equipment, not the equipment itself. Or to put it another way, if a piece equipment is replaced, the new piece of equipment will have the same tag number as the last.

Engineering and IT

Mr Kent said he previously worked in as an electrical engineer before moving to IT. For information managers, 'It is handy to have an engineering and IT background,' he said.

An information manager is something of a translator between engineering and IT, he said. 'They need to understand both languages - almost a broker between the two.'

For an example, if an electrician is going to do some maintenance on a light fitting and doesn't want to shut down the entire electricity network, you'd need to disconnect the specific light circuit and phase. The circuit would need to be uniquely identified on the distribution board so you know you shut down the right one. To put the system together, it is helpful to understand both electrics and how the distribution board works.

Enter as you go long

The best time to enter tag data into the system is when the tag is originally created, he said.

'Engineers will obviously be very busy doing a variety of things, they may not be particularly keen to provide you with the data at these times,' he said.

'But don't let people say to you, we'll deal with that at the end - that is a recipe for disaster you won't be able to get the information at the end,' he said.

'Don't be persuaded to procrastinate on these things you need to collect at the optimum time.'

Buy-in

'You need the whole team to buy in - engineering procurement, operations and maintenance. You need high level management sponsorship. If you get that in place, everyone adopts it, you'll have a successful project and collect all the data that you need.'

'It is important you train your team to use the tools and explain to them why this information is required.'

Content management systems

There are content management system (CMS) companies who can validate your data, and improve the accuracy of it. You can also work with OEM (original equipment manufacturer) documentation and other data sheets.

Two respected companies in the CMS arena are Phusion (formerly Pearson-Harper and ShareCat Solutions.

'By using a 3rd party CMS company, you can ask the suppliers for less data. If you give the basic information, OEM name and part number, they can give a lot of the attributes you require.
You don't have to ask them for 60 to 100 attributes for every tag.'

Working with suppliers

One audience member, from an equipment supplier, asked about what engineering companies are doing to help equipment suppliers do information management.

'We discuss the data handover at the kick-off meetings - explain what is involved with this - give examples of the type of data we are expecting and mechanisms to collect that data,' Mr Kent said.

'One of the dangers we have is that you [as an equipment manufacturer] may not have the right person in the meeting to understand what we're looking for, you might not get representation from information management to understand what you are asking for them. We ask that they bring someone along who has an understanding of this.'

Watch Dave Kent’s talk on video and download slides at
http://www.digitalenergyjournal.com/video/851.aspx

Associated Companies
» WorleyParsons London

CREATE A FREE MEMBERSHIP

To attend our free events, receive our newsletter, and receive the free colour Digital Energy Journal.