Insurance industries issues: contribution to prevention and safety in the financial sector ETH Industry DIALOGUE ON THE FUTURE Risk, Security and Privacy: Is robustness in a globalised society an illusion? 15 Nov 2013, Rüschlikon Ernst Zirngast Risk Engineering Services
Insurance industries issues: contribution to prevention and safety in the financial sector SwissRe Model The Dynamic of Incident causation Human Performance Integrity HPI Assessment Tool Assessment Examples A Strong Case 2
Swiss Re model 3
Comprehensive Safety Management Safety Awareness Programme (SAP) Human Performance Integrity (HPI) Change Management Audit Programme (ChangeMAP) «Lifeware» Safety Culture Interaction Hardware Assessment Matrix (ARADB) Software Safety Management Audit in the process industry (SMAPI) 4
Hardware Design Process Layout Fire protection «Lifeware» Hardware 5 Software
Software Management Procedures Regulations Training Contingency plan Incident investigation Near misses investigation system Follow up on recommendations «Lifeware» Hardware 6 Software
«Lifeware» Humans Working atmosphere Blame free environment Just culture Confidence / Trust Ownership Communication Role models Goals, strategies Priorities, safety first Near misses reporting culture «Lifeware» Hardware 7 Software
«Lifeware» Holes in the layer of protection Example AF447 3 minutes of loss of reliable airspeed indication lead to a loss of a fully serviceable latest-generation aircraft and all aboard 12 minutes later. 1 June 2009 A case of operating outside of the safe operating envelop of humans not of the airplane. «Liveware» keywords Completely surprised Unpredicted behaviours Self generated upsets Failure to recognise what instruments showed them Source Flight International, 20.11. 2012 8
«Lifeware» Holes in the layer of protection Example AF447 From the BEA, the French accident investigator: The word "unpredicted" applied to piloting activity is significant where aircraft design is concerned. When aircraft are first certificated, their design and airworthiness of the airframe and its components are tested, but the certification logic also contains assumptions that the pilots have certain basic competencies. Source Flight International, 20.11. 2012 9
Evolution of Safety Improvements in the process industry Fireproofing Fire fighting Spacing EBV, ESD CM, RCM Hardware ok Flixborough Pasadena Number of Incidents Hardware Software MOC, PTW Training Operating Procedures Emerg. Oper. Proc. Emergency Response Lost time analysis Dissatisfaction with incidents Safety improvements Retrofits Incident invest. 70 s and 80 s Liveware Safety Awareness Programs (Safety Culture) Human Error Safety Behaviour Change Management PSM Root cause analysis Late 80 s Questionnaires, Socio-technical system analysis, Near Miss Report System Mid 90 s Time 10
The dynamic of incident causation 11
The dynamic of incident causation Lifeware defences - Safety culture - Motivation - Alertness Software defences Hardware defences - Process design - Plant layout - Protection systems - Procedures - Audits - Management systems Adapted from Reason, 1990 12
The dynamic of incident causation Unusual conditions create gaps in defences Latent failures in specific systems Software defences Hardware defences - Process design - Plant layout - Protection systems - Procedures - Audits - Management systems Adapted from Reason, 1990 13
The dynamic of incident causation Accident Lifeware defences - Safety culture - Motivation - Alertness Software defences Hardware defences - Process design - Plant layout - Protection systems - Procedures - Audits - Management systems Adapted from Reason, 1990 14
Human Performance Integrity HPI Assessment Tool 15
Assessment tool development in conjunction with ETH and PSI Slide 16
Introduction, HPI issues Team constellation Fatigue Noise Normalisation of deviance Production pressure Shift organisation Stress level Emergency drills Management style Cardian rhythm Availability of EOP, SOP, PSSR Lighting Eating, drinking Fever Skill training level Alarm flooding Zirngast / Kiriakidis Slide 17
Never underestimate the importance of people There is no substitute for humans where creativity, flexibility and random intervention are required ( European Space Agency, 20th Century) Beat Rüegger Slide 18
An analogy Performance integrity of a centrifugal pump Head Safe operating Speed 1 envelope Speed 2 Flow Zirngast / Kiriakidis Slide 19
HPI Holes in the layer of protection Example - Ryanair «Lifeware» Hardware Source Flight International, 16.6.2009 20 Software
HPI Holes in the layer of protection Example - Ryanair «Lifeware» Hardware Source Flight International, 16.6.2009 21 Software
HPI Holes in the layer of protection Example - Ryanair «Lifeware» matters SMS matters 1. Adversarial relationship between pilots & management 1. Can boss see if pilot is stressed and unfit to fly? 2. Young son had just died and not report it to operations 2. Role of copilot, empowerment, CRM? 3. Did not tell that was unfit to fly, due to his emotional state 4. Did fear that death was an invalid reason for being excused to fly 5. Job was insecure Environment matters 1. Thunderstorms in Rome Source Flight International, 16.6.2009 22
CREAM Cognitive Reliability & Error Analysis Method, Hollnagel Erik (Resilience Engineering) 4 main control modes Human Performance reliability High Medium Low Scrambled Opportunistic Tactical Strategic 1.0E-1<p<1.0E-0 1.0E-2<p<0.5E-0 1.0E-3<p<1.0E-1 0.5E-5<p<1.0E-2 Slide 23 Type of control
CREAM 4 main control modes Scrambled (HPI lost, failure probability 0.1 to 1) Choice of next haphazard action, little or no thinking involve, task demand high, loss of situational awareness, momentary panic Opportunistic (HPI random, failure probability 0.01 to 0.5) Choice of action based on present condition, little planning or anticipation, not clear context, constrained time Tactical (HPI controlled, failure probability 0.001 to 0.1) Performance based on planning, follows procedures/rules Strategic (HPI robust, failure probability 0.5E-5 to 0.01) Person considers the global context, wider time horizon, robust performance Zirngast / Kiriakidis Slide 24
CREAM 9 Common Performance conditions (CPC) Adequacy of organization Working conditions MMI & operate. support Procedures & plans Number of simult. goals Available time Time of day (Fatigue) Training & experience Crew collaboration Slide 25
CREAM 9 Common Performance conditions CPCs either improve (+) or reduce (-) or are not significant to (0) human performance Zirngast / Kiriakidis Slide 26
CREAM basic diagram Σ Improved Reliability [7,0] Fatigue & Number of goals cannot Improve performance [Hollangel,1998] 7 6 5 4 3 2 [0,9] 1 Slide 27 1 2 3 4 5 6 7 8 9 Σ Reduced Reliability
Factors correlation Working conditions Fatigue Lighting Noise Temperature Humidity Shift work Age Sleepiness Fitness Distance Illness Work experience Training Crew collaboration MMI & oper. support Availability of procedures/plans Crew cooperation Work satisfaction Leadership Social skills Adequacy of organization Training & experience Equipment External support Number of goals & Available time Decision-Making skills Activity planning Production planning Maintenance planning Operating instructions Available time Stress Slide 28
Example CPC/ Questions Answers Grade Weight Result Adequacy of organisation If the supervisor is unavailable, who takes the responsibility of an emergency situation Who is empowered to shut down an individual unit in an emergency situation? The supervisor ensures all safety requirements are met at work Any operator 0.5 1 0.5 Any operator 1 6 6 Yes 1 3 3 Total result = 9.5 Adequacy of organization Improved Improved Not significant 3<x 7 Reduced >7 Slide 29 3
Example CPC/ Questions Answers Grade Weight Result Adequacy of organisation If the supervisor is unavailable, who takes the responsibility of an emergency situation Who is empowered to shut down an individual unit in an emergency situation? The supervisor ensures all safety requirements are met at work Not defined 0 1 0 Any Operator 1 6 6 No 0 3 Total result = 6 Not significant 0 Adequacy of organization Improved >7Not significant 3<x 7 Reduced Slide 30 3
Results 11 completed questionnaires 11 different plants, 10 different locations Plant assessment with respect to safety A Human Performance Integrity of a plant could be: Robust Controlled Random Lost Zirngast / Kiriakidis Slide 31
Human Performance Integrity Rating on CREAM basis Improve Human Performance 7 6 G 5 4 C I B,F 3 E Europe (A, D, E, F) U.S.A. (J, K) Middle East (B) India (C) Chile (G) Australia (I) Canada (H) Robust Controlled Random Lost A-K: Oil Refineries K J D H 2 A 1 1 2 3 4 5 6 7 8 9 Slide 32 Reduce Human Performance
Human Performance Integrity Rating on CREAM basis Improve Human Performance 7 Robust 6 Controlled 5 Random 4 Lost 3 Doc. Loss 2 1 1 9 2 3 4 5 6 7 Slide 33 8 Reduce Human Performance
Thank you
Legal notice 2013 Swiss Re. All rights reserved. You are not permitted to create any modifications or derivatives of this presentation or to use it for commercial or other public purposes without the prior written permission of Swiss Re. Although all the information used was taken from reliable sources, Swiss Re does not accept any responsibility for the accuracy or comprehensiveness of the details given. All liability for the accuracy and completeness thereof or for any damage resulting from the use of the information contained in this presentation is expressly excluded. Under no circumstances shall Swiss Re or its Group companies be liable for any financial and/or consequential loss relating to this presentation. 35