Collective learningfrom incidents and accidents

Eric Marsden

<eric.marsden@foncsi.org>

First SAF€RA symposium

March 2014, Berlin

What does it mean for an organization to learn?

Organizations have no memory.Only people have memory andthey move on.‘‘ ’’— Trevor Kletz

2 / 20 March 2014 — First SAF€RA symposium

What is collective learning?

Most research on learning focuses on individual cognition

This presentation:

. organizational learning: flow of lessons into new practices andmodified procedures

. policy learning: impact of lessons on public policy, law, regulationsand standards

3 / 20 March 2014 — First SAF€RA symposium

Perspectives on collective learning

learning = processing, interpreting and reacting to information

Research in management science

. action/failure/feedback/correction cycle leading to observableorganizational change

. learning as an organizational aggregate (“as-if” organizationhas cognitive capacity)

. learning of key actors in an organizational setting

. learning emerges from social interactions in the workplace

. the way a group of people act collectively is influenced byshared values & meanings associated with artifacts

technical

organization

=individu

alwritlarge

individu

al=

organization

writsm

all

4 / 20 March 2014 — First SAF€RA symposium

Perspectives on collective learning

learning = processing, interpreting and reacting to information

. action/failure/feedback/correction cycle leading to observableorganizational change

. learning as an organizational aggregate (“as-if” organizationhas cognitive capacity)

. learning of key actors in an organizational setting

Research in psychology

Attractive perspective because it provides an easilymanaged tool for intervention (training targeting singleindividuals)

. learning emerges from social interactions in the workplace

. the way a group of people act collectively is influenced byshared values & meanings associated with artifacts

technical

cognitive

organization

=individu

alwritlarge

individu

al=

organization

writsm

all

4 / 20 March 2014 — First SAF€RA symposium

Perspectives on collective learning

learning = processing, interpreting and reacting to information

. action/failure/feedback/correction cycle leading to observableorganizational change

. learning as an organizational aggregate (“as-if” organizationhas cognitive capacity)

. learning of key actors in an organizational setting

. learning emerges from social interactions in the workplace

. the way a group of people act collectively is influenced byshared values & meanings associated with artifacts

Research in sociology, organization studies

Organizational knowledge is unique to each organization

technical

cognitive

social /cultural

organization

=individu

alwritlarge

individu

al=

organization

writsm

all

4 / 20 March 2014 — First SAF€RA symposium

Perspectives on collective learning

learning = processing, interpreting and reacting to information

. action/failure/feedback/correction cycle leading to observableorganizational change

. learning as an organizational aggregate (“as-if” organizationhas cognitive capacity)

. learning of key actors in an organizational setting

. learning emerges from social interactions in the workplace

. the way a group of people act collectively is influenced byshared values & meanings associated with artifacts

technical

cognitive

social /cultural

organization

=individu

alwritlarge

individu

al=

organization

writsm

all

4 / 20 March 2014 — First SAF€RA symposium

Focus in this presentation

. Learning from incidents and accidents in high hazard industries• process industry, oil & gas, aviation, nuclear, railways

. Focus on organized processes for experience feedback and on informallearning

• and barriers to their success. Focus here on operations, but worthwhile thinking also about learning

potential• during the design phase• during maintenance periods• in emergency response• in decomissioning

. Presentation based on• research funded by FonCSI on experience feedback• ESReDA working group on Dynamic learning from accident investigation

5 / 20 March 2014 — First SAF€RA symposium

Vocabulary: « le retour d’expérience»

French term encompassing:

. incident reporting

. event analysis

. operational experience feedback (nuclear industry)

. lessons learned analysis (US military)

. learning from incidents and accidents

. organizational learning

6 / 20 March 2014 — First SAF€RA symposium

The experience feedback loop

identify incidents,anomalies, accidents

transfer informationto the local manager

classify anomalies, analyze causes,de�ne corrective measures, plan their implementation

manage implementationof corrective measures

communicate lessonslearned to peoplepotentially impacted

change procedures, design,attitudes, safety behaviour, ...

7 / 20 March 2014 — First SAF€RA symposium

Symptoms of failure to learn

. Aspects or types of behaviour of an organization which may suggestthe existence of a “learning disease”

. Can be observed by people• working within the system (review of event-analysis process)• external to the system (accident investigators)

. Help a person recognize “we may be running into symptom λ”

. Point them to possible underlying organizational conditions(pathogens) which may help them understand and improve thesituation

8 / 20 March 2014 — First SAF€RA symposium

Underreporting

Underreporting can be caused by:. a blame culture

. fear that reports will be used in litigation or interpreted in a negativeway in performance assessments

. uncertainty as to scope (which incidents should be reported?)

. insufficient feedback to reporters on lessons learned• leading to demotivation

. perverse incentives which reward people for absence of incidents

. deficiencies in the reporting tool• too complex, inappropriate event typologies…

. management does not promote the importance of incidentreporting

9 / 20 March 2014 — First SAF€RA symposium

Analyses stop at immediate causes (1/2)

. Analyses target immediate causes (technical/behavioural) rather thancontributing factors (organizational)

• “operator error” rather than “excessive production pressure”

. Recommendations target lower-power individuals instead of managers

. Recommendations limited to single-loop learning instead ofdouble-loop learning [Argyris & Schön]

. Instead of multi-level learning, recommendations limited to firmdirectly responsible for hazardous activity

• insufficient consideration of role of regulators, legislative framework,impact of insurers

10 / 20 March 2014 — First SAF€RA symposium

Analyses stop at immediate causes (2/2)

Can be caused by:. insufficient training of the people involved in event analysis

• identification of causal factors• understanding systemic causes of failure in complex systems• training to help identify organizational contributions to accidents

. insufficient time available for in-depth analysis• production is prioritized over safety

. managerial bias towards technical fixes rather than organizationalchanges

• managers may wish to downplay their responsibility in incidents, sodownplay organizational contributions to the event

11 / 20 March 2014 — First SAF€RA symposium

Ineffective follow-up on recommendations

Can be caused by:. insufficient budget or time to implement corrective actions

• production is prioritized over safety• management complacency on safety issues

. lack of ownership of recommendations (no buy-in)

. resistance to change

. inadequate monitoring within the safety management system• missing indicators• insufficient management supervision

. inadequate interfacing with the management of change process

It generally takes years for investigations of major accidents to resultin changes at the system level (typically involving the legal, regulatory,and legislative processes).

12 / 20 March 2014 — First SAF€RA symposium

No evaluation of effectiveness of actions

Consolidation of learning potential of incidents: effectiveness of correctiveactions should be evaluated. did implementation of recommendations really fix the underlying

problem?

Can be caused by:

. political pressure: negative evaluation of effectiveness may be seen asimplicit criticism of person who approved the action

. compliance attitude• checklist mentality: people go through the motions without thinking

about real meaning of their work

. system change can make it difficult to measure effectiveness (isolateeffect of recommendation from that of other changes)

13 / 20 March 2014 — First SAF€RA symposium

No evaluation of effectiveness of actions

Consolidation of learning potential of incidents: effectiveness of correctiveactions should be evaluated. did implementation of recommendations really fix the underlying

problem?

Can be caused by:

. political pressure: negative evaluation of effectiveness may be seen asimplicit criticism of person who approved the action

. compliance attitude• checklist mentality: people go through the motions without thinking

about real meaning of their work

. system change can make it difficult to measure effectiveness (isolateeffect of recommendation from that of other changes)

13 / 20 March 2014 — First SAF€RA symposium

No feedback to operators’ safety models

. Safety of complex systems is assured by people who control theproper functioning, detect anomalies and attempt to correct them

. People have built over time a mental model of the system’s operation,types of failures which might arise, their warning signs and thepossible corrective actions

. If they are not open to new information which challenges their mentalmodels, the learning loop will not be completed

. Can be caused by:• operational staff too busy to reflect on the fundamentals which produce

safety (“production prioritized over safety”)• organizational culture allows people to be overconfident (lack of

questioning attitude)• mistrust of the analysis team• reluctance to accept change in one’s beliefs

14 / 20 March 2014 — First SAF€RA symposium

No feedback to operators’ safety models

. Safety of complex systems is assured by people who control theproper functioning, detect anomalies and attempt to correct them

. People have built over time a mental model of the system’s operation,types of failures which might arise, their warning signs and thepossible corrective actions

. If they are not open to new information which challenges their mentalmodels, the learning loop will not be completed

. Can be caused by:• operational staff too busy to reflect on the fundamentals which produce

safety (“production prioritized over safety”)• organizational culture allows people to be overconfident (lack of

questioning attitude)• mistrust of the analysis team• reluctance to accept change in one’s beliefs

14 / 20 March 2014 — First SAF€RA symposium

Loss of knowledge/expertise

People forget things. Organizations forget things.

Can be caused by:. effects of outsourcing (knowledge is transferred to people outside the

organization). ageing workforce and insufficient knowledge transfer from

experienced workers. insufficient use of knowledge management tools. inadequate or insufficient training. insufficient adaptation (including unlearning), which is necessary to

cope with changing environment/context

Any deviation not properly processed through the reporting systemwill eventually be forgotten!

15 / 20 March 2014 — First SAF€RA symposium

Pathogens

Pathogen: an underlying organizational condition which hinders learningand may lead to one or more symptoms of failure to learn

. Denial (“it couldn’t happen to us”)• related to cognitive dissonance, where people cannot accept the level

of risk to which they are exposed• accident demonstrates that our worldview is incorrect• some fundamental assumptions we made concerning safety of systemwere wrong

• paradigm shifts are very expensive to individuals (since they requirethem to change mental models and beliefs) and take a long time to leadto change

. Resistance to change• trying new ways of doing things is not encouraged• organizations have a low level of intrinsic capacity of change, and often

require endogenous pressure (from the regulator, from changes tolegislation) to evolve

16 / 20 March 2014 — First SAF€RA symposium

Pathogens

. Lack of psychological safety• shared belief within a workgroup that people are able to speak upwithout being ridiculed or sanctioned [Edmondson 1999]

• no topics which team members feel are “taboo”

. Anxiety or fear• anxiety related to legal responsibility, or to loss of prestige• can lead organisations and individuals to become highly defensive

. Drift into failure• organizations gradually reduce their safety margins and take on more

risk over time [Rasmussen & Svedung 2000]

17 / 20 March 2014 — First SAF€RA symposium

Pathogens

1

10

30

600

. Organizational beliefs about safety and safety management, such as:• “improving personal safety improves process safety”

• structuralist interpretation of Bird’s incident/accident pyramid• “if we work enough at eliminating incidents, we will make big accidentsimpossible”

• “rotten apple” model of system safety [Dekker]• “our system would be safe if it were not for a small number of unfocusedindividuals, whom we need to identify and retrain (or remove from thesystem)”

. Inadequate communication

18 / 20 March 2014 — First SAF€RA symposium

Pathogens

. Conflicting messages [Goffmann]• “front-stage”: the actor formally performs and adheres to conventions

that have meaning to the audience• “back-stage”: performers are present but without an audience• disconnect between management’s front-stage slogans concerning safety

and reality of back-stage decisions → loss of credibility

. Pursuit of the wrong kind of excellence• Example: use of incomplete set of KPIs for safety (BP Texas City)• Example: perverse incentives caused by poorly chosen targets for

performance targets

. Ritualization of experience feedback/accident investigation• feeling that safety is ensured when everyone ticks the correct boxes in

their checklists and follows all procedures to the letter• no thought as to the meaning of the procedures

19 / 20 March 2014 — First SAF€RA symposium

Thanks for your attention!

More information on experience feedback at www.foncsi.org

Follow the FonCSI on Twitter: @TheFonCSI

This presentation is distributed under theterms of the Creative Commons Attribution-ShareAlike licence

20 / 20 March 2014 — First SAF€RA symposium