Preparation

The intent of this step is to ensure that the information relevant to the study is gathered, the Chair becomes familiar with this information and they set nodes to allow effective execution of the study. In this phase one of the most important steps is Chair/Facilitator selection.  When selecting a Chair, it is important to consider a few things:

• Duration of HAZOP – some facilitators are quite good at short HAZOPs but don’t have the focus for a longer campaign. In addition, longer campaigns require significant preparation and therefore require a team suitably diligent to complete this. Longer studies can be completed in rotation with good communication between the teams.
• Complexity of HAZOP – Facilitators relatively new to HAZOP may only be comfortable with relatively simple HAZOPs, that would be a HAZOP of a system with 1-2 drawings or a system which has no to little interaction with other systems. HAZOPs with larger numbers of drawings that interrelate with various systems require an experienced facilitator or a strong engineer with operational experience to help lead the team into discovering the interdependencies and the potential hazards.
• Personalities in the HAZOP – If you are arranging a HAZOP with strong personalities who are likely to overpower quieter voices in the HAZOP you need an experienced Chair. This does not necessarily mean that their voice is louder however they have a technique which allows the quieter voices in the room to be heard.
• Time allocated – The time allocated to the scope is often decided before you even engage the Chair however this is critical to your success. So, listen to the Chair if they say they feel the scope will take longer, they have significant experience in this. It is better to send a meeting invite for 5 days and take 4 and a half than to have to reschedule to cover the last half a day.
• HAZOP type – There are several HAZOP types which I will cover in a post later this month however it is important to agree the type of HAZOP you are completing before you start the preparation. Some require significantly less detail than others however they likewise don’t produce the granularity of hazard understanding as the more detailed studies.
• Experience on topic area – Although this is not a definite requirement it can help the flow of your HAZOP if the Chair has experience in your topic area. Although you can mitigate this by bringing in an independent expert in the field to the meeting, this may actually be more useful than having a Chair who once HAZOP’d a system such as yours.
• Independence – The best HAZOPs have someone in the team who is independent of the design or operation team. This can be the Chair (never the scribe) however if this is not possible this can be a process engineer from another area of the business or a technical expert in this particular topic area. It is important to ensure the person has the relevant experience to act independently and to highlight to the person that they are taking this role as it will change their behaviour in the meeting to more challenging interventions as their role is to test if you really have thought of everything.

Once the Chair has been selected, exchange of information will occur.  It is important to give enough time for the documents to be reviewed in preparation for the HAZOP.  This will ensure that the Chair can familiarise with hazards they are not familiar with and select safety moments to focus the team.  If sufficient time is given, they can often come back with a short list of the big-ticket findings before the HAZOP, allowing for correction prior to the meeting to reduce the number of actions.  Normally this type of review should be completed internally with a checklist prior to engaging the Chair, however if expertise or resources are not available, providing additional time to the Chair before the meeting can be a good way of doing this.  During this information exchange provide the Chair with the relevant regional or Company standards to things such as isolations, maintenance preparation, venting flaring and blowdown. Also relevant philosophies and hazard definitions for the Company would be provided however I will cover this in more detail later this month.

The drawings will be split into Nodes. This is normally done from one isolation valve to another to allow complete review of an isolatable system. If these systems are in line with the Quantitative Risk Assessment (QRA) then it makes assessment of the consequences of the hazards easier.

This material will be used to develop Terms of Reference (ToR).  It is important to issue this document to attendees at least one week before the meeting to ensure everyone has the same information about the intent of the meeting.  This ToR can be developed by the Chair, however it is most often developed inhouse alongside the Chair as they define the noding and guidewords.  More details on the contents of ToR can be found in the Standard.

Part of the ToR is the meeting arrangements such as attendees.  Try to keep attendees to 6 or less with required specialist disciplines on call.  Ensure the room is sufficiently sized to accommodate everyone ideally the team will sit around a table facing each other and being able to see the notes projected on the screen with scribe seated next to the Chair.  Only plan to HAZOP for 6 hours a day unless the other people attending have no other responsibilities during this time.  It also works quite well to only HAZOP 4 days a week as it allows time for other pressing issues which can arise in operations and projects. This ensures the team are fully focussed on the meeting when attending.  It also allows a weekly report out by the Chair to management and preparation time for the following week.  If possible the organiser should provide suitable catering for the meeting as this can make a big impact on people’s engagement and energy levels.

During the Meeting

The intent of this phase is for the Chair to facilitate brainstorming with the participants on the various guidewords and deviations, for the Scribe to record these discussions and for participants to engage with the meeting to reduce the number of recommendations (by providing information to the room) and to ensure that all potential hazards are identified and adequately mitigated. Having attended so many HAZOPs and managed the close out of almost all of them participation during the meeting (next to preparation) is the single most important element to managing your hazards effectively. If people are on their phones, emails or reviewing other documents they are not paying attention to the meeting and not adding the value.  The brainstorming can only be effective if the key people are all brainstorming together, a list of key people can be found in the standard however the Chair will help you pull this together.  If there are more than 10 people in the meeting this can discourage quieter members or members with different cultures, languages or from different sites from participating.  Also, it can be difficult to keep one conversation in the room.  Furthermore, when the key participants are distracted the recommendation wording goes out the window resulting in recommendations which solve world hunger and peace.  If you are invited to a HAZOP take the time to find out if you are needed to attend full time, for one node or just on call. It is recommended that all participants (even those on call) attend the introduction presentation which covers the full scope, how a HAZOP works and the timing of the nodes. If you are called upon to participate in the meeting, then participate. What you have to say is useful and may well be significant.

There are two types of recording mentioned in the standard and I will detail them below with the pros and cons:

Full Recording

All the discussions in the HAZOP are recorded even if a recommendation or hazard are not identified.

Pros

• There is a full record of discussion
• If new information arises it is easier to review the notes to see what was discussed
• If a subject was considered and a hazard was not identified the reasoning can be understood (perhaps the inventory was assessed to be too low or storage remote)
• All drawings, revisions, process parameters and attendees can be found from one document for things like incident investigation, revalidation, enforcement etc.

Cons

• It can be more time consuming
• Reports are longer
• It takes diligence in a team

Recording by Exception

Only discussions which result in a recommendation or a hazard are recorded.

Pros

• Meetings flow faster
• Report is shorter
• Can be useful for review of existing systems or procedures (see below)

Cons

• There is a potential that discussions are not allowed to fully develop in the meeting as they initially do not identify a hazard
• If an incident occurs or new information arises unless it was associated with a recorded line it is nearly impossible to understand the thinking of the team
• If new information arises it almost always means a re-HAZOP as insufficient information is available in the report to understand the thinking of the team

It is also possible to record without making recommendations. In my experience full recording, recording recommendations and recording the severity of consequences (not the likelihood) are all good practice in getting a quality report which has longevity. In addition, I add a type of recommendation which is called ‘Amendments’ which are drawing or document errors which do not pose a hazard and therefore do not need to be tracked to closure. What I mean by longevity is that the effort is put in to make sure that all issues are examined and documented which means if this is the asset or operational HAZOP in 5 years a HAZOP by difference can be conducted as you can feel assured that the first round was thorough. This should only be revalidated once and then the next session will be a full HAZOP again.

Regardless of recording type the attendees who attend the meeting should be recorded and the sessions which they recorded and how they attend (part-time, video call etc.). The HAZOP node should have a full description of design intent, design conditions and operating conditions to allow deviations from these to be recorded.

After the Meeting

The intent of this phase is to deliver the recordings of the meeting in a meaningful way which can be utilised by operations or the project team to address the concerns identified without introducing additional hazards. If additional hazards are introduced, then this should be managed with HAZOP by difference or re-HAZOP. 

Hopefully in the ToR you defined the period of time the Chair has to write the report and any special requirements you have for the report such as action sheets. This part should be relatively simple. There is some duplication from the ToR defining who attended the meeting, when it was held and how it was conducted on what system. Then there should be a section for the list of recommendations from the Chair. The HAZOP notes should be included in the report and the HAZOP drawings with clearly marked nodes. It is also very useful if the Chair provides an Executive Summary of the meeting to help summarise the report into other reports such as Safety Case documentation.

Recommendations are named as such to ensure that it is not assumed that the exact action identified in the HAZOP meeting is what is needed to be executed. For example, during the meeting you may identify that a bleed valve did not have a cap on it, so you raised a recommendation to put a cap on valve XY-1234. However after the meeting you realise that all such bleed valves have no cap on it, therefore to meet the intent of the HAZOP meeting (no uncontrolled release from bleed valves to the environment) you may change the recommendation to say that all bleed valves such as valve XY-1234 should be capped and drawing and field to be updated. Or vice versa which would limit the rework to be completed by the action party. You can also decide to not implement a recommendation. 

The recommendation should be clear and concise and should include the justification and intent for the recommendation. For example ‘replace the piston pump with a centrifugal pump’ is not a good recommendation as it does not allow freedom for the engineer to identify a solution to the problem as they do not know the problem and there is potential for error as the pump has not been named. Instead perhaps ‘To reduce the potential for overpressure of the system downstream of pump P101, upstream of valve XY-2345, consider replacing the piston pump with a centrifugal pump to reduce the potential for loss of containment and potential high pressure release directed at the main walkway’. This wording allows the engineer to understand that the issue is loss of containment from overpressure and the location of the walkway relative to the potential loss of containment site. Therefore, there are actually several options here, changing for centrifugal pump, changing the motor on the piston pump to reduce power, installing a barrier between the pump and the walkway or actually uprating the lower pressure pipework. With potentially other solutions I have not identified. This approach also limits the ‘engineering in workshop’ approach that many attendees have which wastes workshop time and doesn’t often come up with the right solution.

An action close-out report should be issued when all actions are closed or at predefined intervals. This should include the justification for not converting recommendations to actions and the action closure to allow a good audit trail. It is often a good idea (especially when there are many actions) to get the original Chair to come back and run a close out workshop to assess whether the action closures have in some way introduced new hazards. The alternative is to wait for the next HAZOP in the project cycle to complete this activity.

Disclaimer: My articles are based on my engineering experience in two major operators over 11 years. They do not reflect the processes of either and are not endorsed by either. If you find an error in the text please feel free to correct me as I too am prone to human error (although I like to think at a lower than average frequency). The intent of these articles is to educate. From beginners to experienced engineers. I hope that through the article or reference material you all get something from it. If you find the content too basic or too advanced again get in touch. Any improvement suggestions welcome.

Kindly reviewed by Jerry Lane, Joâo Abruzzini and Yosef Aljasem