Identifying Performance Influencing Factors

Humans Factors

Introduction

Recently I have posted on Positive Performance Influencing Factors (PIFs), also known as Performance Shaping Factors (PSFs), and what Optimising Performance Influencing Factors means. I seem to have missed out a stage in dealing with PIFs which is the procedure for identifying them as part of the assessment process. Guidance is unclear about what the procedure should be, largely leaving it up to the assessor to use their judgement. I’ve had a couple of discussions with Human Factors practitioners which suggests we could do with something more specific to help people meet the expectations of the regulator. Just to be clear that HSE’s current interpretation of “all measures necessary”, as stated in L111 The Control of Major Accident Hazards Regulations 2015: Guidance on the Regulations, is that companies should reduce risks as low as reasonably practicable (ALARP). To demonstrate ALARP, companies must show that they have complied with relevant good practice (published industry or regulatory standards) and then ask, “’what more can I do to reduce the risks?’ And, secondly, explain: ‘Why have I not done it’” (p.22). I don’t want to get too hung up on interpreting the law in this post, because that’s not the interesting bit, but it’s fair to say that relevant good practice is not clear on the subject of identifying Performance Influencing Factors.

Identification of Performance Influencing Factors

As a starter for 10, I thought I would explain what I expect based on my experience of carrying out and verifying human reliability assessments. Disclaimer: this is my personal opinion.

Firstly, Performance Influencing Factors can loosely be divided into three categories – it doesn’t really matter what you call these but HSE divides them into Job, Person and Organisation. By the way, the list provided by HSE is not fully inclusive; it provides common examples as prompts. The type of PIF will dictate how you assess its influence. Job PIFs such as:

  • Clarity of signs, signals, instructions and other information
  • System/equipment interface
  • Difficulty/complexity of task
  • Working environment

can only really be assessed on plant and during the walk-through. This means that as part of the walk-through/talk-through, the note taker should be listing, perhaps even photographing or videoing, any factors that might influence the task, and how they might influence the task. The reason this has to be done on plant is that people have imperfect memories and the way we ask questions can produce misleading answers. For example, I worked with a company who carried out this part of the assessment in their workshop exercise away from the plant. When asked, “is this valve easy to reach?” The operatives replied, “yes no problem”. Enter inspector. I asked to be shown the valve on plant. It was easy to reach, if you stood on the narrow gauge hydrocarbon pipe immediately below it. I remember vividly how the Process Safety Engineer paled. Context is everything. That’s not job done either. Depending on what the Job PIF is, more work may be required to determine whether it is optimal, good enough or more work is required. If communications are critical, have you checked radio reception is good in all the plant areas?  Is a lighting survey required? As part of your assessment, you should indicate where such a study is required, or why it is not. And throughout all of this, it is important to keep your eye on when the task is likely to be carried out. Time of day effects many environmental factors, as does time of year. I rarely see this recorded in an assessment however.

In my experience, person PIFs are often the hardest Performance Influencing Factors to assess and address. These include factors such as:

  • Physical capability and condition
  • Fatigue (acute from temporary situation, or chronic)
  • Stress/morale
  • Work overload/underload

When assessing whether person factors might influence the task, you are often looking indirectly, simply because individuals differ and there’s really little learning to be gained by asking one or two people how they find the task. I’m not saying don’t ask, but it’s hard to be representative. Some of these will need to be assessed as part of the on plant walk-through/talk-through. Does a task step require physical strength, endurance or flexibility. If so, some people might struggle with it, and if it can’t be redesigned, the company needs to think about how they manage that, especially with our ageing workforce. Are there task steps that require complex analysis and/or rapid decision-making and/or passive monitoring – fatigue is going to be an issue if it’s a 24/7 operation. Likewise, if in an emergency, the task may last more than 12 hours. What else might the operatives be doing at peak times of workload. If the normal is unloading one tanker, how will they cope with three?

Stress, morale, competence, motivation – these could affect the whole of the task – I privately call them global PIFs – but the situation now may not be the situation in a year’s time. In other words, organisational change can threaten a deterioration in all of these (or promise improvement, she writes optimistically). Organisation Factors also tend to be global PIFs. These include factors such as:

  • Culture
  • Supervision
  • Leadership
  • Clarity of roles and responsibilities

So for global PIFs, whether Person or Organisation, my assumption is that these will usually affect performance across the whole task. I personally see little point in repetitively listing these against every task step in the HRA record, unless they might disproportionately affect a particular task step. A particularly difficult task step might rely heavily on the competence of the individual, for example, whilst skills for other steps are transferable from other tasks i.e. this one is different and difficult. It might also really require supervision too. My personal opinion is, that for many of the Organisational PIFs, an organisational approach is required. Whilst leadership may well affect the likelihood of someone breaking the rules (for better or worse), it does not exist in relation to just one task. Leadership needs to be consistent across work activities to engender trust. Likewise, building a just safety culture can take years. Here I believe we are better off looking to see that the company has policies and procedures and programmes to deal with these Performance Influencing Factors than trying to assess them in a Human Reliability Assessment.

So, that is my starter for 10, and I am very interested in hearing views from you. Please post a response!