Abstract
The aim of the meeting was to provide an opportunity for presentations and discussion on the influence of human factors on radiation safety. The roles and responsibilities of both organisations and individuals were considered together with the requirements for training. Appropriate methods were discussed for continuing improvement, learning from experience and formal accreditation.
The keynote address was given by Geoff Webb (IAEA) who identified the `seven deadly sins' of irresponsibility, carelessness, ignorance, tolerance, wilfulness, arrogance and avarice with examples from industrial irradiation facilities, radiotherapy and medical radiography to show how human factors could compromise radiation safety.
The first session covered organisation and cultural factors. Dick Taylor (Magnox Electric) began by exploring the difficult question `What is Safety Culture?' and by examining a number of different views. Safety culture can be considered to be a combination of a safety management framework and staff attitude at all levels. Ideally within an organisation there should be a questioning attitude, a rigorous and prudent approach to safety issues and good communication of safety related information. Organisations which have a good safety culture should actively seek information, welcome new ideas and encourage the sharing of responsibility. Dick also identified some typical barriers to the establishment of a safety culture; in particular, lack of management commitment, complacency and inadequate monitoring.
Philip Dendy (Addenbrooks Hospital) examined management responsibilities by looking at ICRP 73 `Radiation Protection and Safety in Medicine' and taking some half dozen management ideas from it to illustrate how they apply in a large teaching hospital. Recent initiatives such as the enhancement of nurses' roles in authorising x-rays have led to the development of training/induction courses which address, amongst other things, safety protocol. Other training includes guidance for doctors in how best to use a clinical radiology department, general POPUMET training and the role and membership of a radiation safety committee.
Alex Rankine (UKAEA) was able to draw on recent first hand experience of the effects of organisational change on safety issues. UKAEA has changed its focus over the years from an original aim of developing the uses of nuclear energy and associated technology towards its current objective of safely dismantling redundant nuclear facilities. An important factor affecting safety is the relationships between different organisations.
The presenter then moved on to focus on specific areas of radiological interest. There was some discussion on dose reduction on UKAEA sites where total site doses have decreased by almost 90% since 1986. Organisational change has probably focused attention on dose budgets more than in the past. Another consequence of recent organisational change is the increasing number of RPAs having a role to play in the overall radiological protection arrangements. UKAEA, managing agencies, main contractors and sub-contractors might all have different RPAs but the UKAEA view is that its RPA should be `more equal than the others' in his role representing the interests of the site licensee.
In the same way, cooperation between approved dosimetry services is important when considering internal dosimetry and some consideration has been given to the desirability of insisting on the use of a single ADS for all internal dosimetry purposes.
In summary, operating a nuclear licensed site with contractors and tenants does present some challenges in the coordination of overall radiological protection arrangements. Under such circumstances cooperation between the various parties becomes paramount.
To start the next session on current practice and issues Jerry Williams (HSE) presented a series of examples showing how human factors affect reliability. Much of this presentation came across as being a distillation of common sense such as the observations that a ten fold reduction in human error can reduce risk by a factor of 100 and that a trained operator made typically 15 times less errors than the naive worker. Another example demonstrated the enhanced reactions to combined auditory and visual stimuli compared with those for either one or the other. Jerry, nonetheless, gave us all a very useful reminder of factors to consider in risk reduction and dose control.
On the subject of operational reviews and benchmarking Nick Schroeter (Magnox Electric) described the plant evaluation (or peer review) process applied to the nuclear power stations operated by Nuclear Electric, Magnox Electric, Scottish Nuclear and BNFL. The process is used to promote continuous improvement in station safety and reliability. Nick explained how the process operates and the benefits derived by the participating companies.
Patricia Wright (MRC) gave a valuable insight into the evaluation of operating procedures and the way they are presented to the user. She maintained that good procedures should provide answers to the readers' questions and that if the reader doesn't have questions then he/she is unlikely to persevere in reading a particular document. Typical questions would be of the What?, When?, Why?, How? variety.
Procedures should be able to be used to specify target information within the procedure and search for it. It should also be possible to skim through the document to pick up key information. The wording needs to be quite explicit to optimise understanding and interpretation by the user. In particular it must be clear to the user where knowledge needs to be applied to make a decision.
Troubleshooting of procedures needs to address not only the formal aspects of procedure layout and presentation but also the empirical aspects of how the task is performed. Ian Dalling (AEAT) stressed the importance of safety culture within organisations and went on to describe a safety performance model which he had produced in an attempt to promote a clearer understanding of the organisational components that are critical to safety. Research has indicated that, within organisations, a better safety culture correlates with a good safety performance. A company building a positive safety culture encourages individuals and teams at all levels to be responsive and proactive in their approach to safety issues. The adoption of common principles and definitions for the structure of organisational safety has the potential to accelerate the growth in the understanding of enhancing safety performance.
On the second day the meeting focused on Training and Development. Wendy Bines (HSE) gave a regulator's view. Having referred to the requirement under the Management of Health and Safety Regulations 1992 for `adequate health and safety training' Wendy also discussed the provision of `refresher training' and the demonstration of `competence'. Training can be grouped under three main headings: organisation, job and individual (the latter generally being linked to performance appraisal). In order to ensure adequate training there are several key steps which need to be followed such as deciding if training is necessary, identifying training needs and objectives, carrying out the training and evaluating its effectiveness subsequently. Competence can be demonstrated by experience under supervision, by accreditation, through NVQs and SVQs, and by licensing (e.g. Corgi gas fitters). Reference was also made to EC Directive 96/29/Euratom which is to be implemented by 13 May 2000 and which addresses information, training and qualified experts relevant to radiological protection. These factors are also being considered by HSE's IRRs Revision Topic Group No. 5 `Organisation for Exposure Restriction'.
No forum discussing radiological protection training would be complete without a contribution from Monty Guest (University College, Suffolk). Monty reiterated the basic aims of any form of training which are to transfer specific knowledge, skills, ideas and attitudes to the trainee. Particular emphasis was placed on the open learning options now available. Also highlighted was the desirability of ensuring that the learner is involved in decisions related to his training, thereby ensuring learner motivation, ownership and responsibility. Objectives need to be set which are both demonstrable and testable. Self-checking as the learner progresses through the course can be useful. Quality standards are an important aspect of any training and can be attained through validation, moderation and monitoring.
The role of the Mentor is becoming increasingly important in modern training programmes. Mentors can help the learner by overcoming obstacles in the learning process and by ensuring an appreciation of learning in practical work situations. The Mentor role is very much to meet, supervise, assess, help, listen, correct and occasionally take action. It is always important to note that the more a Mentor does on behalf of the learner, the less the learner learns.
Dave Drury (Nuclear Electric, Sizewell B) gave an account of how his organisation carries out radiation protection skill broadening for Operations and Maintenance staff. Sizewell `B' has developed the concept of a `radworker' whose prime role is not health physics but who is trained to carry out radiation and contamination monitoring duties within clearly defined limits. As radworkers the scope of their duties is to carry out monitoring at their own work locations, to monitor tools and equipment leaving controlled areas and to establish work area boundaries. A specially designed radworker training course has been implemented which is of 60 hours duration (2 weeks) with a biennial refresher course of 3 - 5 days duration. This compares with the initial 4 week health physics monitors' course. It comprises both practical and theoretical modules with a strong emphasis on the practical application of radiological protection measures. After a series of practical and written assessments suitable individuals are given final authorisation by an RPA.
In practice the Health Physics Section at Sizewell provides instrumentation, clothing and equipment for use by radworkers while offering guidance and advice when required and undertaking audit monitoring to check that acceptable standards are being maintained. The overall lessons learned at Sizewell are that there need to be clearly defined terms of reference for the radworkers, commitment at all supervisory levels and continuing health physics support. In setting up such a system consideration needs to be given to feedback from the various work groups which may be affected.
Richard Lerski (Ninewells Hospital, Dundee) outlined in some detail the computer assisted learning package (CALRAD) developed at Dundee to address the training requirements associated with the Protection of Persons Undergoing Medical Examination or Treatment Regulations 1988 (POPUMET). He discussed the advantages of computer assisted learning (CAL) in general and how it was particularly suited to radiological protection and had proved to be effective. Although a certain degree of computer literacy was needed by students, feedback had indicated that 60% preferred CAL to conventional methods.
Since retiring from the HSE Peter Beaver has chaired the SRP Qualifications and Professional Standards Committee which has been a major player in the development of vocational qualifications in radiation protection which are about to become available to radiation protection practitioners and support staff.
Peter outlined the NVQ system as applied to radiation protection. Vocational qualifications are national, involve wide coverage of fields of work, are transferable and form part of a hierarchy of qualifications. The lead body is the Occupational Health and Safety Board and the assessing body the City & Guilds of London Institute. Assessors may be in-house or external and advice is currently coming from a reconstituted Technical Assessment Group. Assessment is based on a combination of portfolio of evidence and interview. Registration and assessment costs are likely to be of the order of £400 - 600 at Levels 2 - 4.
Reference was also made to the SRP Certification scheme for experienced radiation protection practitioners. Presently this is renewable on a 5-yearly basis, based on a portfolio of evidence and it is hoped to tie it into a continuing professional development (CPD) scheme which should be up and running soon. It is based on defined fields of work, it is relatively low cost and it is a tripartite scheme which has been extended beyond the confines of the SRP membership.
The meeting then turned its attention towards accidents and the lessons to be learned from them. Nick Pidgeon (University of Wales, Bangor) outlined current `Man Made Disasters Theory'. This suggests that human and organisational causes make up 70% - 80% of background preconditions to disasters, that failures are socio-economic and that preconditions multiply over time (incubation period + trigger event). There are different behaviour types which may be involved; slips/lapses/mistakes, small group (as opposed to large organisations) and organisational/management (large organisations may have widespread faults). Complex interactions are involved, often between apparently separate items.
There are two main routes for developing learning opportunities. During the incubation period learning can be blocked by information difficulties. After a disaster the problems are of blame and organisational politics. Imagination has to be used to address safety issues during a potential incubation period, including attempting to fear the worst, asking the question `What if?' and generally making worst case assumptions. Social systems need to be developed which promote a `no-blame' culture and ask `how can candid reporting be encouraged?'.
Good feedback systems help users to learn lessons from incidents and to improve safety. This was the tenor of the next presentation by John Croft (NRPB). It is also of use to `the Authorities' in determining regulatory requirements, enforcement strategy and emergency planning arrangements. Problems with feedback can include distortion before it gets to the people who need to use it, depth of coverage (lack of detail) and availability.
IAEA feedback in the form of accident reports aim to identify why an accident happened and the associated circumstances. Medical management of casualties is also examined. Furthermore its `Lessons learned ...' reports can serve as useful training material.
Recently the HSE, Environment Agency and NRPB have developed the Ionising Radiations Incident Database (IRID). This looks at accidents involving occupational and public exposure but does not include nuclear incidents, transport incidents or patient exposures on which information is gathered elsewhere. NRPB is administering the IRID system and would welcome inputs from sources throughout the radiation protection profession. All information is unattributable and confidential. In due course NRPB intends to compile case histories and slide sets for training use.
Mark Kearney (Nuclear Electric) examined the information systems that exist for providing feedback. Nuclear Electric's Central Feedback Unit interfaces with many other systems and organisations. It serves all UK reactor operators and has the dual objectives of maintaining compliance with Site Licence condition No. 7 and minimising adverse events to maximise company profitability. There are also targets to reduce unplanned discharges and to reduce the unplanned reactor trip rate from typically 3/4 per year for AGRs to a world average of one per year for other reactor types.
The Central Feedback Unit includes the Nuclear Plant Event Report (NUPER) and Nuclear Good Practices Information Database (NUGID) systems. It also provides feedback in a variety of useful forms such as training videos, newsletters or face-to-face presentations and can include event information, management information and safety expenditure justifications.
The final presentation took the form of a comparison of the Chernobyl and Bhopal Incidents by Malcolm Grimston (Imperial College). Both incidents have been blamed on human error but there are significant differences between the two events. In the case of Chernobyl there was no component failure involved and only one clearly inadvertent operator action. At Bhopal there was only component failure with no clearly inadvertent operator actions.
The culture at Chernobyl was such that short cuts were taken as a matter of routine but pressures on operators were such that they felt unable to discuss issues or concerns with colleagues or regulators. To some extent there was also a widespread belief that a major accident was impossible.
At Bhopal there was generally low morale. At Chernobyl there was high morale but complacency.
Credit is due to Chris Perks who organised the programme and speakers for this very successful meeting which stimulated a considerable amount of lively debate during the open fora. Some 200 delegates attended over the two days.