A colleague described a very unusual situation in his project. A facility was undergoing an upgrade on the inlet of the facility. Nothing unusual about that, but what was unusual was as a result of the upgrade, 100% of the feed would flow through the new upgrade. And … while that statement by itself is not unusual, what was unusual was the transition. There would be a time when 100% of the feed HAD to flow through the new system, while it was being commissioned and before it had been handed over to operations.
Let me say that differently … an entire complex was forced to rely on an “incomplete” system to continue to operate. And I am not talking about partial capacity; I am talking about 100% of the feed having to go through an incomplete system before entering the rest of the complex. If the new system was not available, the rest of the complex would get no feed.
As you can see, I am taking liberties with the word “incomplete”, because an uncommissioned system has the potential to work and work well, but it is not ready to be given to operations. It has not been thoroughly tested, and the potential for spurious trips is quite high.
The team was able to minimise the exposure, but not totally eliminate it. The activities done to minimise the exposure included:
+ Temporary piping modifications
+ Procedural modifications
+ Alternative commissioning and testing fluids
+ Operating the system in 100% recycle independent of the main plant
+ And other smaller methods
This is clearly not a desirable situation. While there are other circumstances that make the project execution plan of THIS project unique, I wondered what could have been done to avoid this problem. Here are some of my thoughts:
Expand the risk assessments. The risk assessment for this project was well done, but was done on a completed basis (looking at the change after implementation). The risk assessment did not look at the method or process of change. While it is very easy to say this is a constructability issue, I must disagree – I do not know of any organisation that will willingly surrender complete control of their process flow unless there was no alternative. Therefore, the risk assessments (such as HAZOP) should include commissioning plans and procedures during the HAZOP assessment. This will increase the scope of the HAZOP type reviews (and therefore the cost and duration), but that is the risk of risk assessments … more scope = more cost = more potential benefit.
Provide dedicated commissioning reviews during design. This will be difficult to do because of a lack of quality commissioning personnel – commissioning people are commissioning, not in a review – but again consider the alternatives. It is desired to commission with as many things working as practicable – and not operating in manual. Operating in manual increases risk.
Ensure commissioning addresses the transition from construction to commissioning AND commissioning to operations. Often these are overlapped, and I have observed that organisations tend to do one very well, but for some reason seem to have trouble doing both very well. The handover process involves multiple parties and teams – final handover to operations is just one of many handovers.
I have intentionally not specifically mentioned pre-commissioning. It further complicates the story, but does not change the basic story.
My colleague has advised me that this project has been a mixture of success and problems, but so far the problems have been managed through a variety of reasons. I hope that continues – he is a good person and deserves a good outcome.