Case Study in Successful Risk Management – The Oil and Gas Industry in Norway Some critical reflections Terje Aven University of Stavanger, Norway

Regulatory development – a thorny path or a freeway Command and control Self regulation What regulative approach should we choose?

The regulations give general requirements related to Risk assessments Risk acceptance criteria Risk reduction processes o ALARP: Risk shall be reduced to a level that is As Low As Reasonably Practicable

Challenge Industry risk perspective is rather narrow and compliance oriented The regulator has a broader perspective

”Optimal decision” Risk analysis Cost-benefit analyses Risk acceptance criteria Tolerability limits

Decision Science Management Ethics, politics

”The formula ” risk analysis

P =1 x Risk descriptionWhat is acceptable risk This approach should be used with care !

Risk acceptance criteria P < 0.1% P < 0.01%

Take risk Weight given to E Reduce the risks and uncertainties Cautionary-precautionary E[NPV], cost-benefit analyses Balance Development and protection Risk acceptance criteria E: Expected value

RISK is more than computed probabilities

Challenge Industry risk perspective is rather narrow and compliance oriented The regulator has a broader perspective

Probability-based Historical data Probability-based Historical data Knowledge dimension + + Surprises Risk perspective Industry practice

Kahneman asserts that we have a basic lack of ability to treat small risks: we either ignore them completely or give them too much weight. The main thesis put forward is that we grossly over-estimate small risks. Kahneman’s book «Thinking fast and slow»

For Kahneman the risk was acceptable

I knew that the risk was truly negligible, and that any effect at all on my actions would assign an inordinately high “decision weight” to a minuscule probability. In fact, I was more likely to be injured in a driving accident than by stopping near a bus. But my avoidance of buses was not motivated by a rational concern for survival.

+ + Risk perspective

Petroleum Safety Authority Norway

Main problems with the probability based approach

John offers you a game: throwing a die ”1,2,3,4,5”: 6 ”6”: -24 What is your risk?

Risk (C,P): 6 5/ /6 Is based on an important assumption – the die is fair

Assumption 1: … Assumption 2: … Assumption 3: … Assumption 4: … … Assumption 50: The platform jacket structure will withstand a ship collision energy of 14 MJ Assumption 51: There will be no hot work on the platform Assumption 52: The work permit system is adhered to Assumption 53: The reliability of the blowdown system is p Assumption 54: There will be N crane lifts per year … Assumption 100: … … “Background knowledge” Model: A very crude gas dispersion model is applied

+ + Risk perspective

Black Swan A surprising, extreme event relative to the present knowledge/beliefs Aven (2013) On the meaning of a black swan in a risk context. Safety Science, 57, 44-51

Black swans Extreme consequences a) Unknown unknowns b) Unknown knowns c) Known but not believed to occur because of low judged probability

Aven (2014) Risk, surprises and black swans Routledge

Challenge Industry risk perspective is rather narrow and compliance oriented The regulator has a broader perspective Implications for the implementation of the ALARP principle

Risk shall be reduced to a level that is As Low As Reasonably Practicable (ALARP) Costs Benefits A risk reducing measure should be implemented provided it cannot be demonstrated that the costs are grossly disproportionate relative to the gains obtained The burden of proof is reversed. ALARP

Take risk Weight given to E Reduce the risks and uncertainties Cautionary-precautionary E[NPV], cost-benefit analyses Balance Development and protection Risk acceptance criteria E: Expected value ALARP

Regulatory development – a thorny path or a freeway Command and control Self regulation What regulative approach should we choose?

Challenge Industry risk perspective is rather narrow and compliance oriented The regulator has a broader perspective

35 A new perspective on how to understand, assess and manage risk and the unforeseen Project Norwegian Oil and Gas Association: Preventing Major Accidents Transfer of experience, building knowledge and expertise

Extra

Hansson and Aven (2014)

Risk description (A,C,U) Q: Measure of uncertainty (e.g. P) K: Background knowledge C’: Specific consequences (C,U) C’ Q K

Probability Consequence, impact Poor background knowledge Strong background knowledge Medium strong background knowledge

Cautionary principle Faced with risks and uncertainties caution shall be a ruling principle Not execute an activity Robustness Defence in depth Redundancy ALARP Precaution …

Precautionary principle If there is scientific uncertainty related to the consequences …

Risk analysis Risk picture Acceptable risk Unacceptable risk All agree

A surprise for some Not a surprise for others Unknown knowns

The Deepwater Horizon accident  Erroneous assessments of pressure test results  Failure to identify formation fluid penetrating the well despite log data showing that this was happening  The diverter system was unable to divert gas  The cutting valve (blind shear ram – BSR) in the BOP failed to seal the well og-anbefalinger-for-norsk-petroleumsvirksomhet-article html

Disasters We experience combinations of conditions and events which collectively result in a major accident We do not normally identify such combinations of conditions and events in risk assessments – and, if we did, they would typically be disregarded because of negligible probability 45 A B C D

How to confront the black swans Knowledge Risk assessments

Modified procedure for using RAC If risk is found acceptable according to probability with large margins, the risk is judged as acceptable unless the strength of knowledge is weak (in this case the probability-based approach should not be given much weight). If risk is found acceptable according to probability, and the strength of knowledge is strong, the risk is judged as acceptable. If risk is found acceptable according to probability with moderate or small margins, and the strength of knowledge is not strong, the risk is judged as unacceptable and measures are required to reduce risk. If risk is found unacceptable according to probability, the risk is judged as unacceptable and measures are required to reduce risk.