Turnaround Efficiently: Some Tips for Shutdown Season
Skeleton maintenance crews during Covid-19 will mean a tsunami of maintenance activity when Planned Plant Shutdowns are due. How can you maximise maintenance efficiency?
With the covid-19 restrictions, many companies have been reducing their maintenance staff to a minimum and the maintenance activities to essential to reduce people’s exposure to the disease.
However, as time moves on the maintenance that has not been done will rise in urgency and is likely to have a surge, especially when restrictions ease and when one or more of the pending vaccinations become available.
The surge is due to the plant carrying or patching-up low-level faults and patched or postponed maintenance will be growing in risk as time and usage move on. Eventually, this will lead to a surge in urgent maintenance work. This means excellent and highly efficient work preparation will be essential to manage this surge and ensure risk remains within acceptable limits.
The various Continuous Improvement methodologies can be deployed to streamline work preparation activities to improve concurrent operations and to ensure every working hour is utilised in the most valuable way. This is were eliminating as much waste in the work preparation and execution processes as possible will provide real short and long term business value.
In the next few months, we will be in the Offshore industry’s shutdown season, this is the time where various offshore facilities will be shutting down production for essential planned maintenance known as “Turn-A-Rounds” or TARs. One of the most disruptive shutdowns is the Forties Pipeline System (FPS). There are a large number of wells and facilities that rely on this pipeline and its ancillary facilities to operate. Years in advance the operating company of the FPS, BP, make public when its shutdown will occur and for how long. From this commitment, the operators and maintainers of the facilities that rely on the FPS plan their own TARs. These windows often last a matter of weeks even as short as two weeks, this leaves very little leeway and error room, so companies put a huge amount of effort in getting as much work as possible done during this time.
That is not to say they don’t get it wrong, I have witnessed times where the word has gone out that a shutdown is in the pipeline (if you’ll excuse the metaphor) and every engineering team creates a plan, a list of work to fill the available time, without taking the other teams into account meaning plans many times the number of available manhours is created with only a percentage having any chance of being carried out. So, even if the unused plans are enacted at a later date these would have to be revisited for an update before they were useable, pure waste.
To utilise the time and workspace as efficiently as possible work planning needs to fit the shutdown opportunity be as complete and accurate as possible and everything needs to be where required just-in-time for the work to take place.
The operators and maintenance companies often put in place a gate planning process for their normal planned maintenance. The planning for work begins 12 months from the due date. At this point the details are fuzzy often is no more than what type of maintenance needs to be done and a due date. But the scheduling and maintenance teams will already be working out what maintenance work can be carried out concurrently. As the due date draws closer, the maintenance and scheduling teams will be adding definition to the work, what types of job need to be done and budgets. The maintenance team will be working out how the work will be executed and from that deciding the requirements for the job these usually fall into the following categories: People, Equipment, Materials, Spares and Documentation.
There will be gate checks in place, where a designated gatekeeper will check the state of the work planning against set criteria for that particular gate, e.g. contractors notified, equipment ordered, spares purchased. Different companies will have different gate intervals often the last gate will be 1 or 2 weeks from the due date and after this, the schedule is frozen. Often it is expected all the inanimate elements will be required to be in position, on the worksite. The people will have been notified of where they are to be and, if necessary, transportation and accommodation booked.
So, if this work planning process is in place why is there difficulty? Let’s discuss a few typical difficulties:
Barely meeting Gate Compliance
Often there is pressure to get the work planned by the gate target date, as with anything problems arise and the work planner does the bare minimum to get their work plan through the gate review meaning to catch-up later on the detail. As with any job, their attention gets drawn away by the next most urgent job and they don’t think about the one they have just passed through the gate until it becomes urgent. This is where the gatekeeper must be decisive, if the job is not up to the spirit of all the planning criteria for that gate it must be pushed back to a later due date. There are two reasons for this, the job is likely to be in the same position in the next gate review, and it sends a message to the work planners that they will not be able to pass the work through the gate with just the bare minimum of planning, setting the expected standards for detail, the detail that will ensure the job can go ahead.
I worked on a major project where the only metrics for preparing the work-packs was lead times and quantity of work packs completed by certain dates, as incentives these made sure the work-packs were “completed” on time but didn’t ensure.
Discovering work after the shutdown starts
There is an unknown element with every job and example to illustrate this; I was renovating my house and wanted to reroute the wires of a light. I shut off the power and cut an access hole so I could push my hand through, what I discovered was the cables wouldn’t pull through, I had to cut another access hole only to find the cables had been clipped to the rafter which was stopping it from pulling through cleanly; the lead time for the job expanded when I opened up to see what the job entailed. This was a small job, imagine on a TAR that a pipeline inspection showed significant corrosion through-out the whole section of piping and all the work planner’s data modelling suggested only a 3rd of the pipeline would need corrosion treatment, the plan lead time for that job needs to extend by up to 60%, not to mention the knock-on effect to other concurrent and subsequent work.
Some suggestions to cope with emergent work:
- Plan to open everything on day one and reassess the work plans to adjust for discoverable, or emergent, work early in the shutdown. Rescheduling will obviously need to be done, but this will be at the beginning of the TAR meaning much less schedule change later.
- Keep a corporate experience library for the work planners to refer to when working out lead times for jobs. As the work comes to completion archive the plans and the planned vs actual timings. Now mandate consulting the library as part of the work planning process perhaps even add it in as a gate criterion.
- Do a practice TAR. An Asian pharmaceutical firm carried out a practice TAR before the shutdown, the resulting activities worked perfectly, everyone knew what they needed to do, all the equipment and materials were confirmed to be in place and the documentations was proven. Even the emergent work had much less impact on the schedule than on previous TARs.
When running a TAR, meetings are essential for decision making and for agreeing and tracking actions. Often project and functional managers’ calendars are crammed with mandated meetings and these meetings are about updating or reviewing stuff. Can attendees update and look at the performance before the meeting starts? Perhaps using visual management?
What happens when someone is late to a meeting? Everyone else’s time is wasted. Meeting discipline is essential, plan the meeting, ensure there is an agenda, back it up with visual management, agree on meeting rules among the team (all visual management needs to be updated before the meeting starts). End the meeting on time (if meetings regularly overrun either the meeting needs to be extended, the scope needs to be reduced or the team needs to be more disciplined in keeping to the agenda). One person should run the meeting, this means challenging when people are going down rabbit holes. If anyone is starting to glaze this is a symptom that the individual’s time is being wasted and that the meeting is being taken down a rabbit hole. The meeting owner needs to curtail the discussion and direct it to be completed outside the meeting). The leader must make decisions on what to do if someone doesn’t show on time, which will include contacting them to discuss how to prevent a recurrence. The leader should make sure the meeting ends on time so the meeting discipline can be paid forward.
Planning with old data and models
I have spoken to work schedulers who complain that no one talks to them, these were the same people who told me that part of planning is weekly schedule re-baselines (completely re-estimating the schedule). Work schedulers need to be contacting the work planners and they both need to have a relationship with the technicians, that way when the work plans are created and then formed into a schedule, the planning and scheduling models and software have a chance to be accurate, issues can be understood and dealt with before significant disruption to the schedule.
Another way to improve work planning and scheduling is to build updating the planning and scheduling systems after into the close-out of each job. If a planning system like Primavera or Maximo is used, ensure the algorithms and planning assumptions are updated regularly. One organisation I worked with hadn’t updated their planning models and assumptions in years, accepting that it was just part of the process that the schedule needed to be adjusted many times a day, even during the frozen schedule.
Differences between the system and reality
Often work planners pass their jobs through the gate criteria based-upon what they see on the logistics tracking system (e.g. SAP) but how often are these wrong. Work planners, before submitting jobs to the final planning gate should be physically able to put their hands on all the materiel required before hitting the button to commit the job to the frozen schedule. In Aberdeen, a major energy company built a huge warehouse to reduce the amount of transportation required. They set that warehouse up so, as items arrived, they were put with the rest of the materiel for that job, many of the work planners were based in that building, meaning it was easy for them to walk down the stairs to the dedicated space and put their hands on the kit before hitting the commit button.
Too much stuff
This issue covers over-ordering, losing stuff and not removing used stuff quickly. Ordering more than is necessary just in case, which mightn’t be a problem on an individual level, but if every job had this fat deliberately planned in it would eat up the budget to no value as well as fill-up storage space. I worked with an oil platform team that had to take a month to purge the platform of unnecessary equipment (including shipping containers) as it was near its safe working weight limit. It would have been cheaper and easier to evacuate outstanding materiel at the end of each job.
If the stuff required for a job is sent to the worksite in dribs and drabs (e.g. as soon as each vendor can supply them) soon the site stores will be full and the stores' team will have to get creative; perhaps storing things in corridors or under stairs. Then, when the work planner commits the job (because the computer says everything is present) the technicians arrive, and if even one thing is missing the best case is the job is delayed, worst, the job is postponed and the technicians have to leave the site and come back to do the job at a later date. To prevent this issue stage the equipment and materials, and only ship them when everything is there and the job is ready to execute. This way when the materiel arrives at the work area it is all there and is easy for the technician to find.
Often I have seen jobs either being cancelled or the workers having to be demobilised from the site because their beds are needed for some break-in work. They are told just to grab their personal kit and get on the transport, and that their equipment will be sent on. On top of cluttering the worksite, often this kit is on hire, and the customer is paying that rental. So, keep track of any kit coming on to the site make sure it has a demobilisation date and ensure it is removed back to the vendor by that date. I ran a project in a major energy company, they had just transitioned from one tracking system to another, all the equipment that was on the platform during that transition dropped off the system and was essentially invisible. This meant there was kit on the platform for months with the customer paying charges when they weren’t required, in one example we found three test kits from three different vendors each charging different hire rates, stored on a shelf in the electrical workshop, the supervisor believing that they belonged to the company. By the time we caught it the company had paid £48k and this was increasing £500 per day.
In another project we needed space, so we began looking asking about the shipping containers stored around the area. Some of them could be accounted for and were supporting ongoing or scheduled work, but 10% of them weren’t required for any work, and when we as some of the technicians, they said they didn’t know when they arrived, they were there for as long as they could remember. When we looked in some of them, they were empty, another reason why speedy disposal is as important as on-time delivery.
So, coming into TAR season there are things that can be done to reduce lead times, improve planning accuracy and keep to, even reduce, budgets; initially to cope with the maintenance surge but latterly the lessons can become the new way of working spreading the benefits into the long term… it is all down to great planning.
Darren Clyde has spent 15 years working with organisations and teams to reduce cost and effort of doing business. He has worked with many teams in different countries from a wide range of backgrounds including, aviation, military, and energy organizations to reduce the business impacts of maintenance, both planned and unplanned.