You are Home   »   News   »   View Article

Measurables better than probabilities for risk

Friday, December 2, 2011

The risk of offshore operations is better looked at in terms of direct measurables - do we have a barrier or not - rather than probabilities, says Dr Robin Pitblado, director for HSE Risk Management Services at DNV

Dr Robin Pitblado, director for HSE Risk Management Services for risk management consultancy DNV Energy said that during operations it is not very helpful to look at a risk in terms of a probability, but better to look at it in terms of direct measurables - whether or not a barrier is failed or degraded, and how to make sure that the barrier doesn't fail.

'Knowledge of the status of barriers is the key,' he said. 'In order to have an incident, several of your barriers must have failed.'

He was speaking at the 'Integrated Operations' conference in Trondheim, Norway, on September 13-14, organised by the Trondheim Center for Integrated Operations in the Petroleum Industry.

DNV performed many of the 'HAZOP's' (hazard and operability studies) of all the processes which took place after the Macondo disaster, including flares, vessels and risers. This was to ensure safety for the response teams and to ensure no action made things worse.

One approach is to continually monitor the overall risk in terms of barrier status. So for example if a control system barrier has a problem, you can examine if the overall risk of the system goes up or stays the same - and respond accordingly.

You can make rules of what operations are 'permitted' - i.e. allowed to occur - in the event of certain barriers being degraded. For example, you can determine whether or not it should be allowed to move a deepwater rig if the GPS is not available.

Dr Pitblado said that there needs to be more focus on process safety, not just occupational safety.

The oil and gas industry overall has improved occupational safety by 6-10 times over the last 20 years, but there has not been the same improvement in process safety. 'Neither the EU nor the USA have been able to reduce major accidents,' he said.

The key findings of the different studies into Deepwater Horizon / Macondo disaster were that 'the accident was preventable, mistakes were made,' he said. BP, in its investigation, identified 8 barriers that failed, any one of which should have prevented the scale of event.

DNV was commissioned to do the forensic investigation of the blowout preventer. While DNV cannot speculate beyond the report for legal reasons, this study showed that it failed to close because the high pressure, high volume production fluids going through the middle of the BOP at the time pushed the drill pipe (inside the BOP) to the side, where the knife could not cut it, he said.

One exception is the UK North Sea, where 'major leaks have reduced from around 19 per year in 1996 to about 2 now,' he said.

DNV believes that the industry worldwide can improve major accident safety 'by a factor of 10, or in the North Sea by a factor of 3,' he said.

DNV does consulting work helping companies improve process safety, by looking at the consequences of different process safety disasters, the different barriers, how the barriers are being monitored, human performance, and how often various events happen.

Integrated Operations can help, by making it easier to introduce new shore based specialists. If the offshore team is more confident than they should be given the current risk situation, the 'shore-based people should not have that mindset,' he said. 'They can ensure corporate safety standards are not weakened by local pressures - financial, personal KPI's, etc.'
Onshore staff can also verify that procedures are being followed fully and correctly, activity is within the permitted envelope, and barriers are functional.

However nobody has yet worked out how to utilize an integrated operations centre which could work in an emergency situation, but in any case, 'in the last 10 minutes [before an explosion] maybe IO is not the way to solve the problems,' he said. 'Many problems develop in the weeks leading up to an incident and IO would be very helpful for these.' Crisis centers after an event are well established.

On the question of whether the oil and gas industry is more complex than other industries, Dr Pitblado said that the oil and gas industry can be simpler in some ways, in that it is easier to understand whether or not you have a problem.

'If our safety systems work we can reduce the likelihood and amount of the leak or mitigate it after release. In the Texas City accident, there was material going into a column and no material coming out where it should, so it was going somewhere else and that led to the large release' he said.

'Other industries have more complexities and things you don't know and greater 3rd party interference. For example, consider the railway industry, with problems with trespassers on the track and many other things you don't know or control.'



BOX TEXT
You can view the complete presentations with video plus slides online at www.ioconf.no click on 'presentations 2011' and 'video 2011', then look for 'Plenum 2'.



comments powered by Disqus

CREATE A FREE MEMBERSHIP

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

DIGITAL ENERGY JOURNAL

Latest Edition June July 2018
Jun 2018

Download latest and back issues

COMPANIES SUPPORTING ONE OR MORE DIGITAL ENERGY JOURNAL EVENTS INCLUDE

Learn more about supporting Digital Energy Journal