Unlocking safety culture excellence: our behaviour is the key

Transcription

1 Unocking safety cuture exceence: our behaviour is the key John Hunter 1 and Ronny Lardner 2 1 EHS Leader, GaxoSmithKine, Irvine, UK 2 Chartered Psychoogist, The Kei Centre Ltd, Edinburgh, UK INTRODUCTION The individua s roe in deveoping a strong safety cuture This paper describes the deveopment and depoyment of methods to promote the behaviours which support a strong and sustainabe safety cuture. Most organizations in hazardous industries have embraced the need for a strong safety cuture, and recognize that exceent safety eadership, effective supervision, and high eves of workforce invovement are essentia safety cuture ingredients (Fin, R et a, 2000; HSE, 1999; HSE, 2001). To support the deveopment of a strong safety cuture, both site and topic-based approaches have been adopted. This project used an aternative approach. Site-based approaches Site-based approaches typicay invove some form of safety cuture diagnosis, and a pan to address areas for improvement. The unit of anaysis is the site or organization. The improvement pan typicay incudes a need for change in behaviours and practices at different eves of the organization. Topic-based approaches Topic-based approaches invove interventions to address specific aspects of safety cuture for exampe supervisor or safety eadership deveopment programmes, practices designed to encourage and promote workforce invovement such as appointing workforce safety representatives, or impementing a behavioura safety programme. An aternative approach Whist the site or topic-based approaches are appropriate in some circumstances, they do not describe a the individua behaviours required to deveop and support a strong safety cuture, or specify how these behaviours reate to each other and are mutuay supportive across different eves of the organization. Furthermore, site or topic-based approaches do not aways end themseves to integration into the organisation s existing safety management system or human resources systems. 93

2 GaxoSmithKine s (GSK) Irvine site, and the background to this project The site was estabished in 1973, covers 135 acres, and is ocated on the West coast of Scotand, approximatey 30 mies from Gasgow. Irvine is one of GSK s argest primary manufacturing sites, with 650 staff invoved in the production of both peniciin-based antibiotics and active pharmaceutica intermediates. The approach to improving the safety cuture at GSK Irvine, and the specific project described here are miestones on a journey for the site that commenced back in During that year the Factory Safety Committee had been discussing safety cuture at severa meetings, and utimatey agreed to make the first step in the journey by attempting to measure the existing cuture. In October 2005 as part of the European Heath and Safety Week, a site survey was aunched using the existing HSE Cimate Survey Too (HSE, 1997). Over 400 empoyees (60% of the workforce) competed the 72-question survey. In eary 2006 the Factory Safety Committee then anaysed the survey, with the aim of identifying some key interventions and deveoping supporting action pans. In March 2006, just after ceebrating success at reaching 1 miion hours Lost Time Injury-Free for the first time in the site s 33-year history, a process safety incident occurred when a 4500 itre reactor vesse expoded, bady injuring 2 operators. The site s top priority ever since has been an effective response to prevent any simiar incident recurring, This project has been fundamenta to achieving that goa. After the exposion, the subsequent GSK interna investigation focused argey on the technica aspects of the root cause anaysis of what happened within the reactor vesse. The investigation aso examined some of the EHS behavioura issues concerning staff at a eves invoved directy or indirecty with the incident. This ed to a cear recommendation to review the behavioura findings from the incident in conjunction with the recentycompeted site safety survey. To ead this project, the company appointed an experienced EHS manager (the first author). At this stage it was decided to bring in some independent, externa expertise and the second author was engaged to assist with the anaysis of both the 2005 HSE cimate survey and the behavioura aspects of the March 2006 incident investigation. The second author has been invoved in the subsequent stages of the design and deveopment of the EHS Behaviour Standard and associated interventions, whist GSK staff have presented a training and communication sessions. It was decided to deveop a competency mode which described (a) the specific manageria, supervisory and workforce behaviours which supported exceent EHS performance and (b) those which detracted from exceent performance. DEVELOPING THE COMPETENCY MODEL Five main sources of data were used to deveop this competency mode:- 1. Existing academic research, which identifyed eadership behaviours which support workpace safety outcomes (HSE, 2003) 94

3 2. Existing industry research conducted by the UK offshore sector s cross-industry Step-Change in Safety group, who deveoped a set of safety behaviours foowing a review of 11 offshore fataities (Step-Change in Safety, 2004) 3. Previous work (Hayes et a, 2007) generousy shared with GSK by Wood Group Engineering North Sea Ltd 4. Key EHS behaviours reevant to becoming a high-reiabiity organization (Weick, 1999) were integrated into the mode 5. In-company research, to identify specific positive and negative HSE behaviours which had particuar reevance to recent exposion, and the findings of the incident investigation. It was decided to base the competency mode on the behaviours which differentiate those who are more effective at managing heath and safety, from those who are ess effective. There is an important distinction between the technica competences necessary to do a job (i.e. the abiity to drive a fork-ift truck), and the persona competencies which differentiate between those who are more or ess effective in a job. Athough a group of fork-ift drivers may a possess the same technica competence, individua differences wi exist in how effective they are in achieving their overa job objectives (e.g. safety, housekeeping, efficiency). Tabe 1 beow summarizes the differences between behavioura competencies and technica competences, and how to anayse jobs to derive the behavioura competencies which support superior job performance. In this project, the first job anaysis method used was critica incident interviewing, (Fanagan, 1954) which asks interviewees to identify critica incidents they have persona knowedge of, which ed to a good or poor resut. In this case the resut referred to HSE performance. Incident does not mean accident or oss, it coud simpy be someone s behaviour in a meeting which supported or undermined heath and safety. The second method was repertory grid technique (Key, 1955) which eicits the constructs or attributes which experienced peope use to differentiate between good and Tabe 1. Differences between behavioura competencies and technica competences Behavioura competencies Technica competences Focus Peope who do the job Jobs or tasks which peope do Leve of Performance Superior performance Minimum standard Outputs Behaviours which contribute to superior performance Key roes and tasks. Minimum knowedge, skis and Appropriate job anaysis methods Critica incident technique Repertory grids Studying documentation Structured job anaysis questionnaire Observation 95 abiities required Studying documentation Observation Functiona job anaysis

4 poor job performers. The technique asked experienced peope to think about managers, supervisors or others they know we, and who are (a) effective in managing heath and safety or (b) ess effective or ineffective. By comparing those in groups (a) and (b), it is possibe to define the behaviour(s) which differentiate the effective and ess effective performers. These two job anaysis techniques were used to extract the specific positive and negative EHS behaviours from interviews and focus groups hed with managers, supervisors and technicians. The next steps in mode deveopment were to examine the critica incident and repertory grid data, sort the positive and negative behaviours into reated groups, and differentiate whether they were behaviours required of everyone on the workforce, ony supervisors, or ony managers. The wording of behaviours was refined, to aid carity. Four sets of positive and negative behaviours were identified for each eve in the organization: everyone, supervisors and managers. Each set of behaviours was given a short descriptive abe. The resuting overa scheme is shown in Figure 1 beow. The tota number of behaviours generated across a four sets and three eves totaed 100. Figure 1 iustrates how it is ony when the appropriate behaviours are dispayed by a peope in the organization that an exceent EHS resut can be achieved. This approach Figure 1. Overa scheme of EHS Behaviour Standard 96

5 can be contrasted with many behavioura safety programmes, which focus argey on workforce behaviours. The 12 sets of behaviours were further examined, and it became apparent that four common themes ran through the sets of behaviours: standards; communication; risk management and invovement. These themes emerged from the data gathered, and were not pre-determined. Figure 2 iustrates how the sets of behaviours reate to the four themes. By reading across these themes, it is possibe to see the mutuay-supportive interreationships between the sets of behaviours for each group. Simiary, it is possibe to identify how a ack of the correct behaviours at any eve can undermine the overa resut. For exampe, management efforts to set standards, and workforce efforts to compy can be undermined by the wrong supervisor behaviours. Figure 3 beow is an exampe of the content of one of the tweve sets of behaviours, and shows the positive and negative behavioura indicators which were derived from the job anaysis. The resuting competency mode, known in this company as the EHS Behaviour Standard, has the foowing important features: research-based simpe to understand Figure 2. How EHS Behaviours reate to topics 97

6 Figure 3. Manager s Set High Standards behaviours, with positive and negative indicators defines the positive and negative behaviours which contribute to exceent and poor EHS performance shows the inter-reationship between behaviours of managers, supervisors and everyone in the workforce incudes anguage and exampes which are company and site-specific can be used by individuas and teams to understand their roe in deveoping a strong safety cuture format can be readiy integrated into the organisation s safety management and human resource systems, e.g. induction, seection, training & deveopment, and appraisa DEPLOYMENT Deveoping supporting materias for the EHS Behaviour Standard Materias were prepared for the communications cascade using oca graphic designers and printers. There were 3 key pubications:- 1. A 24-page A5 sized booket. This booket provided an introduction from the Site Director, an overview of the approach to the standards and then sections on each of the 4 sets of behaviours for each of the 3 groupings of staff. The booket aso 98

7 incuded 2 pages at the back providing a process, with worked exampes, as to how to assess onesef against the behaviour standard and to create an individua improvement pan. 2. A sma pocket-sized foding Zip-Card containing summary information on the approach. This pubication sti however contained a 100 behaviours 3. Postcards The intention of the postcard was to encourage staff to indicate one strength and one area for improvement from the EHS Behaviours. These were recorded on the postcard at the communication sessions and posted to the Site Director. This enabed some anaysis to be carried out on common themes and, more importanty, provided some foow-up as the postcards were passed back to the individuas directy via their team eaders 3 months after the training session they attended. Impementing the EHS Behaviour Standard: Re-induction The design of the impementation was informed by (a) the principes of ABC anaysis, a behavioura change technique which emphasises the important of providing consequences to change behaviour (see HSE, 2002 for detais), and (b) the knowedge that it was managers and supervisors behaviour which had to change first, foowed by the workforce. It was decided to re-induct the entire workforce about the EHS Behaviour Standard. For such a significant re-induction programme to be successfu, securing management and supervisory ownership and invovement was critica. To that end, the first EHS Behaviour Standard re-induction sessions were carried out with that target audience. In addition, because of the subject matter and due to the site being unionised with two unions at shop-foor eve, union Safety Representatives were aso invited to the initia sessions. Two haf-day re-induction sessions were deivered on consecutive days in January 2007 with a tota attendance of 83 staff. The content was the same as used for subsequent workforce re-induction sessions with some additiona emphasis on the key responsibiities that management, supervisors and safety reps hed with respect to safety performance and safety cuture. Furthermore, the foowing individua objectives were communicated to this group of staff:- Managers objectives Personay set an exceent exampe of the EHS behaviours at a times Ensure a team members participate in the EHS behaviours re-induction Active monitoring and sponsorship of team action pans arising from re-induction Ensure during Q a managers, supervisors and safety reps compete 360-degree feedback on EHS behaviours, and that any identified deveopment needs are buit into their own PDP Supervisors objectives Personay set an exceent exampe of the EHS behaviours at a times Take an active roe in deivering the EHS behaviours re-induction 99

8 Deveop and support a team action pan on improving EHS behaviours arising from re-induction During Q personay obtain 360 degree feedback on EHS behaviours, identify any deveopment needs and buid into persona PDP To further support this group of key staff a set of Questions and Answers were prepared in advance by the project team and these were handed out at the end of the re-induction sessions. Thereafter haf-day re-induction sessions for the whoe workforce then commenced some two weeks ater. A series of ten sessions were set up over an intensive two-week period. Tota attendance was exceent, with 657 attendees covering more than 95% of site staff. One key target set by the Site Director was to compete the re-induction programme before the first anniversary of the exposion this was successfuy competed with the ast of the ten sessions hed on the 28 February Each re-induction session covered the foowing content, most of which was deivered as team-based activities, ed by oca managers and supervisors An introduction to the subject matter of safety cuture An emphasis on the need for improvement at GSK, Irvine. This invoved some commentary on the two injured operators from 2006 incuding their support for the approach to the EHS Behaviour Standard. Aso the Site Director or a deputy recaed his persona account of the incident from 2nd March These persona messages were very powerfu in what became an emotionay-charged atmosphere. An overview of the CAPA (Corrective And Preventative Action) Pans that were being progressed foowing the incident A simpe worked based scenario to get staff thinking about their EHS behaviours An expanation of how the team created the Irvine EHS Behaviour Standard. This incuded reminding staff of their input via the 2005 cimate survey An overview of the EHS Behaviour Standard and a wak-through the booket The creation of Team Action Pans using a tempate suppied Discussions about foow-up supporting activities Individua commitments recorded using the booket and the postcards. These re-induction sessions worked extremey we because the team eaders were primed to faciitate discussion, and aso because attendees were grouped into work teams to create their own team action pan. Furthermore, at the end of each session, time was aowed for staff to review their own eves of performance against the appropriate set of behaviours and commit to an individua area for improvement. These were captured in the bookets that they took away with them and on postcards which were sent to the Site Director. Foow-up activities to ensure sustained behaviour change Severa foow-up activities were announced at the re-induction sessions to emphasise that the pan was to ensure this new approach to improving our management of safety was not going to be a one-off event. The commitment was there from the eadership team to ensure 100

9 this became fuy embedded in ways of working i.e. strengthening our safety cuture. Foow-up activities incuded:- Individua Commitment captured on postcards at the re-induction sessions, individuay reviewed by Site Manager, who added a persona message, and returned 3 months ater to the individuas via their Team Leader with a discussion about how they are progressing. Team Action Pans deveoped during the workforce re-induction sessions, coated site-wide to check for consistency, and monitored for progress via departmenta and persona objective tracking systems. Manager and Team Leader Objectives as described above each manager and team eader has a set of specific objectives in reation to the EHS Behaviour Standards incuded in their 2007 PDP forms. Checking understanding and soiciting feedback from the workforce re-induction sessions. This was managed via a web-based survey form that was e-maied to a attendees within 2 weeks of their training session. The questionnaire asked for responses to 7 questions to check understanding, 2 questions to assess their eve of support for the programme and then optiona extras to provide genera feedback comments and to submit an idea for consideration. 575 staff had responded with 223 ideas submitted and 149 comments provided. A high eve of understanding of the ro-out sessions, and support for the site s EHS Behaviour Standard was verified via the survey. Winning prize for the best idea heped encourage submissions and embed and sustain the programme. Induction Programme a 30 minute interactive introduction to the EHS Behaviour Standard has been introduced into the site induction programme to ensure a new empoyees are covered as soon as they join. Contractors it was aways the intention to ensure that a contractors based on site were treated in the same way as GSK staff. A contractors (over 300) attended simiar re-induction sessions in Spring Pubicity a number of banners and posters carrying the overview of the Behaviour Standards have been created and are on prominent dispay across the site. Recognition use of the company reward programme has been encouraged to recognise any particuar exceent exampes of the correct dispay of EHS Behaviours. A number of these have aready been awarded on site since the ro-out programme commenced. 360 Degree Feedback this programme for a managers and supervisors commenced with the site eadership team. Each person seects severa coeagues/peers and severa subordinates who together with their ine manager assess the individua against each of the EHS Behaviour Standard Manager behaviours, using a web-based survey too. Supporting consutants then anayse the returns and create a report that is fed back directy to the individua who is then assisted or coached into creating a more detaied individua improvement pan. The target is that a Managers, Supervisors and Safety Reps wi compete this process in

10 Human Factors Training As a COMAH site there is an expectation by the reguator that systems are in pace to manage human factors in reation to our major accident hazard instaations. To date two Human Factors Awareness courses have been hed on site for a cross section of staff from the Factory Safety Committee. Aso a more detaied two day Human Factors Anaysis Toos course (Lardner and Scaife, 2006) was run on site to enabe experienced incident investigators to (a) better understand why peope invoved in incidents behaved as they did, and (b) write better behavioura recommendations which wi infuence behaviour of those immediatey invoved, and others, in the future. RESULTS TO DATE There is supporting evidence of benefits starting to be reaised on a number of EHS KPIs. At the end of September 2007 the foowing highights were noted as the best eves of EHS performance in the site s 33 year history:- Lowest ever recorded 12-month roing LTIIR at Highest-ever recorded number of hours worked since the ast ost time accident 1.45 miion. Passing 1 miion hours LTI free for ony second time ever and in successive years (2006 and 2007). Lowest ever cumuative Spi Index a key environmenta measure. A cross-section of experienced managers, supervisors, EHS professionas and safety representatives used the Safety Cuture Maturity Mode (Kei Centre, 2007) to assess whether and how the site s existing safety cuture had changed since the exposion in In their opinion, the site had improved by one eve, and this was aided by the deveopment and depoyment of the EHS Behaviour Standard. At an eary stage in the project, the site HR Director expressed significant reservations about the abiity of the site, aided by the second author, to achieve a cutura change. In Juy 2007 he commented ii was sceptica about the possibiity of cuture change ii thought it was expensive consutant mumbo-jumbo but I was wrong I take my hat off to you there has been a rea change for the better over the past few months. This project has been recognized internay, as it won the GSK Chief Executive Officer s EHS Award for 2007 in the EHS Initiative category. A tota of 86 projects from GSK sites across 30 countries were submitted and reviewed by a pane of externa judges. The GSK Vice-President and Corporate Head of EHS commented: The efforts represented by this award appication are not ony critica to EHS performance at Irvine, but I beieve that Irvine has set the mark for other sites to foow. It is a great effort and I m very peased that the externa pane judged it so highy. Congratuations and we done to everyone at Irvine. 1 Safety Cuture Maturity is a registered trademark of The Kei Centre Ltd. 102

11 The approach described is now being adopted for use by GSK gobay, as its preferred method of assessing and enhancing safety cuture. DISCUSSION AND CONCLUSIONS The question has been asked If the EHS Behaviour Standard and methodoogy had been in pace a year earier, is it ikey this woud have prevented the reactor exposion?. Foowing the exposion, the incident investigation concuded that a hoistic response was required, invoving improvements to engineering, the safety management system and the existing safety cuture. Using Reason s Swiss Cheese anaogy, there were hoes in mutipe ayers of system defences. The site identified 11 improvement areas from technica issues (Emergency Vent Design, Mechanica Integrity) to management systems (Permit to Work, Change Management) and safety cuture and behaviours. Since then the approach has been to cose a gaps in these system defences (or ayers of cheese). In the opinion of the site management, without the focus on safety cuture and behaviours, the chances of sustaining the benefits across the other improvement areas woud be imited. It is impossibe to say with any certainty whether the EHS Behaviour Standard and methodoogy woud have prevented the exposion, without the engineering and safety management system improvements. The ikeihood woud certainy have been reduced, via management behaviours which pace safety before production, supervisory behaviours which encourage wariness amongst their team members, and everyone being more invoved in EHS, and speaking up about any concerns. An additiona issue addressed very effectivey by the EHS Behaviour Standard and methodoogy was the beief amongst some site empoyees that the deficiencies eading to the exposion were ony reevant to the buiding where the exposion occurred. During deveopment of the EHS Behaviour Standard it became apparent this was not the case, and this misconception was firmy rebutted during the workforce briefings. This nove project heped a site deveop and impement a safety cuture improvement project, in the aftermath of a serious exposion. The methods used proved fexibe, and acceptabe to the target audience. A comprehensive impementation pan was deveoped and executed, with a strong emphasis on shaping manager and supervisor behaviours, as they in turn can strongy infuence the wider workforce. The fexibe nature of the EHS Behaviour standard aowed to be buit into the safety management system, so that its infuence can endure It is ikey that this approach wi be of interest to other organisations ooking for practica methods to deveop and embed the safety cuture they desire. REFERENCES Fanagan, J. C. (1954) The Critica Incident Technique in Psychoogica Buetin, 51, Fin, R., Mearns, K., O Connor, P. & Bryden, R. (2000). Safety cimate: Identifying the common features in Safety Science, 34,

12 Hayes, A et a (2007) Personaising Safety Cuture: What does it mean for me? Paper presented at Loss Prevention 2007, Edinburgh, UK May 2007 HSE (1997) Cimate Survey Too: HSE Books HSE (1999) Effective supervisory safety eadership behaviours in offshore oi and gas industry OTO 0065/1999 HSE Books HSE (2001) Invoving empoyees in heath and safety: Forming partnerships in the chemica industry HSG 217 ISBN Heath and Safety Executive (2002). Strategies to Promote Safe Behaviour as Part of a Heath and Safety Management System. CRR 430/2002, HSE Books: Norwich UK. HSE (2003) The roe of manageria eadership in determining workpace safety outcomes Key, G (1955) The psychoogy of persona constructs New York: Norton Kei Centre (2007) maturity%20mode.htm Lardner, R and Scaife, R (2006) Heping engineers to anayse and infuence the human factors in accidents at work in Process Safety and Environmenta Protection, 84(B3): Step-Change in Safety (2004) Fataity Report Weick, K., Sutciffe, K & Obstfed, D. (1999) Organizing for High Reiabiity: processes of coective mindfuness, in Research in Organizationa Behaviour, 21,