Process Safety Performance Indicators - The UK Experience in ...

23
Process Safety Performance Indicators: The UK Experience in Major Hazard Industries Seveso Conference 2010 Stockholm 19 May 2010 Peter Dawson Principal Process Safety Specialist Inspector Hazardous Installations Directorate UK Health and Safety Executive [email protected]

Transcript of Process Safety Performance Indicators - The UK Experience in ...

Page 1: Process Safety Performance Indicators - The UK Experience in ...

Process Safety Performance Indicators:The UK Experience in Major Hazard Industries

Seveso Conference 2010Stockholm19 May 2010

Peter Dawson

Principal Process Safety Specialist Inspector

Hazardous Installations Directorate

UK Health and Safety [email protected]

Page 2: Process Safety Performance Indicators - The UK Experience in ...

Presentation outline

• Terminology and definitions

• Where we started and why, in the UK.

• Wider perspective - world incident learnings.

• How PSPIs can help prevent major incidents.

• How PSPIs can be developed and used.

• UK approach to PSPI implementation.

• PSPI example

Page 3: Process Safety Performance Indicators - The UK Experience in ...

Terminology & definitions

• Process Safety (PS) – prevention, control and mitigation of incidents and events with potential to cause death, major injuries and/or significant damage. Clear distinction from personal safety.

• Key Performance Indicator (KPI) – includes wide range of metrics used to measure business, operational and safety performance.

• Process Safety Performance Indicator (PSPI) -HSE uses PSPI to mean a small number of selected site specific indicators for monitoring the performance of key risk controls. Specific type of safety KPI.

Page 4: Process Safety Performance Indicators - The UK Experience in ...

Where we started and why in the UK

• BP Refinery, Grangemouth, Scotland 2000: A series of loss of containment incidents –prosecution and a record fine.http://www.hse.gov.uk/comah/bpgrange/

• A good safety record when viewed by lost time incident rate but BP did not know how well it was managing major hazard risks

• Management system discouraged reporting of bad news – senior managers assumption ‘no news is good news’

Page 5: Process Safety Performance Indicators - The UK Experience in ...

Where we started and why in the UK

Subsequent enquiries showed that senior managers of similar sites had:

• An overwhelming belief in system design and integrity and in the regulatory controls –CoMAH/Seveso

• No means of discovering deterioration in the risk controls in place, short of catastrophic failure

• Little or no use of KPIs for major hazard risk and where process safety KPIs were used they were exclusively ‘lagging’.

• An over reliance on auditing that focused on compliance – not safety and environmental outcomes.

Page 6: Process Safety Performance Indicators - The UK Experience in ...

World incident learnings

• Investigation of chemical and process industry major incidents worldwide continue to reveal similar failings.

• Despite widespread communication of the lessons incidents keep occurring in the same ways.

• Andrew Hopkins’ book ‘Failure to learn’, about the BP Texas City disaster provides a detailed analysis of why the company had failed to learn the lessons from earlier incidents, including BP Grangemouth and Exxon Longford.

• Concludes that a major factor was ‘catastrophic risk blindness’

Page 7: Process Safety Performance Indicators - The UK Experience in ...

World incident learnings

Catastrophic risk blindness can arise from;

• focusing too much on personal safety –using lost-time accident rates to measure safety

• poor understanding of causes of major incidents – different from personal safety

• failure to identify and learn from process upsets & deviations – no data collected.

• inadequate Process Safety leadership by senior management

Page 8: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can help

A well designed system of PSPIs can;

• provide specific information on how well process safety risks are being controlled – ‘amplifying’existing ‘weak signals’ – before a major incident.

• challenge ‘no news is good news’ attitude to process safety

• allow improvements in process safety controls to be demonstrated

• provide both ‘leading’ and ‘lagging’ data for maximum benefit

Page 9: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can help

Lagging indicators designed to identify failures of safeguards& controls that can cause upsets or near misses

Operating Procedures

High pressure alarm/trip

Pressure relief

Plant integrity inspection

Safeguarddefects

Possible lagging indicators

Possible leading indicators

Initiating event

Major Accident

Upset

Leading indicators designed to identify weaknesses in key elements of safeguards & controls that could lead to failure

Page 10: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

HSE Guidance HSG 254• Developed jointly with CIA

and individual companies

• Originally started with post BP Grangemouth pilot in Scotland 2003/04

• Step by step guide developed and trialled

• HSG 254 published 2006

• Clear methodology for developing PSPIs linked to MA events for a site.

• Available on HSE website.http://www.hse.gov.uk/pubns/books/hsg254.

htm

Page 11: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

OECD Guidance on Developing Safety Performance Indicators related to Chemical Accident Prevention, Preparedness and Response.

• Produced by working group on chemical accidents

• Recently revised 2nd edition, based on pilot programme with expert review

• Uses HSG 254 methodology

• Sets out 7-step process with 3 examples, more practical detail and explanation than HSG 254

• Freely available on-line at http://www.oecd.org/dataoecd/60/39/21568440.pdf

Page 12: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

New guidance from American Petroleum Institute

• API RP 754 “PSPIs for refining and petrochemical industries”

• Defines four tiers of indicators to provide both leading and lagging data

• Tiers 1 & 2 are loss of containment events against defined threshold levels

• Tiers 3 & 4 provide information on the performance of safety and management systems.

http://www.api.org/standards/psstandards/

Page 13: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

Key starting questions for every organisation:

• How will the information be used? By whom and when?

• Who is involved in setting the indicators?

• What will change in the organisation as a result?

• Do the indicators match the risk profile of the business?

• Have they been prioritised based on vulnerability to deterioration and the relative risk that the control measure protects against?

• Do you measure at a sufficient frequency to detect rapid change?

Page 14: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

Lagging & Leading

• Causes confusion - the most important issue is to obtain the right information

• Set lagging indicators to show critical deviations from the desired outcomes – failures in risk controls and safeguards leading to upset or ‘near miss’.

• All adverse findings must be followed up – indicator doesn’t tell you what the problem is!

• Set leading indicators for the critical ‘must do’activities/elements of controls/safeguards – show defects or weaknesses in advance of a failure.

Page 15: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

Site specific v corporate indicators

• Site based indicators more closely match the risks specific to the processes and activities on site.

• It is easier to involve the workforce in setting site indicators compared to corporate – more relevant.

• Great amount of benefit comes from the analysis required to set indicators – a lot is learnt about the importance of the various control measures.

• Corporate indicators are more suited to benchmark performance across a number of businesses

• Corporate indicators are more relevant where the same risks and systems for controlling them exist across businesses

Page 16: Process Safety Performance Indicators - The UK Experience in ...

How PSPIs can be developed

PS Leadership is vitally important to ensure:

• Process safety is given the right degree of attention and focus;

• Process safety considerations feature in key business decisions, and

• Understanding of major hazard risk and the importance of critical control measures is communicated and championed.

Recent UK PSLG Guidance on PS Leadership Principleshttp://www.hse.gov.uk/comah/buncefield/pslgprinciples.pdf

Page 17: Process Safety Performance Indicators - The UK Experience in ...

UK approach to PSPI implementation

• Senior management and employees should be involved in setting indicators

• Indicators should be set following an analysis of the vulnerability of control measures

• The indicators adopted should match the risk profile of the enterprise

• A balance of leading and lagging indicators should be used

• A small number of focused indicators should be used to avoid ‘overload’.

• CEOs and senior managers should make business decisions taking account of information from indicators

• Information from indicators should be used to improve performance,

• Indicators should be reviewed regularly and changed with experience

Page 18: Process Safety Performance Indicators - The UK Experience in ...

UK approach to PSPI implementation

• HSE expectation for all CoMAH/Seveso sites to have suitable PSPIs in place.

• Implementation requires staged approach to identify, trial and fully implement indicators.

• Top tier sites should have PSPIs by 2011

• Lower tier sites 1-2 years later.

• Series of workshops held for operators and trade associations.

• Progress monitored at site inspections.

Page 19: Process Safety Performance Indicators - The UK Experience in ...

PSPI example

PSPIs for a fuel storage depot with pipeline and jetty filling;

• Buncefield, UK, incident 2005 – large vapour cloud explosion arising from overfilling of petrol storage tank.

• Led to establishment of Process Safety Leadership Group (PSLG)

• Final report “Safety and environmental standards for fuel storage sites” includes worked example of PSPIs in Annex 1 http://www.hse.gov.uk/comah/buncefield/fuel-storage-sites.pdf

Page 20: Process Safety Performance Indicators - The UK Experience in ...

PSPI example – identify major accident causes and risk controls

Control of contractors

Plant change

PTW

Design

Inspection and maintenance

Competence

Operational procedures

Control andinstrumentation

SubsidencePhysical damageWearCorrosionOver-pressureAccidental leakageOverfilling

Challenges to integrityRisk control systems

Page 21: Process Safety Performance Indicators - The UK Experience in ...

PSPI example – lagging indicators

• For each risk control define purpose, or ‘what success looks like’.

• Identify a measurable successful outcome to provide a lagging indicator

• For example;– To prevent overpressure of transfer pipeline – ‘number

of times pressure >10bar during transfer’

– To prevent overfilling of tank – ‘number of times tank filled above defined safe fill level’

Page 22: Process Safety Performance Indicators - The UK Experience in ...

PSPI example – leading indicators

• Identify critical elements or activities of risk controls, need to consider which

– Must work correctly every time

– Are more likely to deteriorate over time

– Are undertaken most frequently

• For example;– To prevent overpressure – number of times ship unloaded without ship to shore checks correctly completed

– To prevent overfilling - % completion of inspections and test of tank gauging system

Page 23: Process Safety Performance Indicators - The UK Experience in ...

Thank you for listening

Any questions?