Optimising maintenance resources
Maintenance strategy is one of the main pillars for sustaining production availability in an oil rig or refinery.
In my experience, when developing a RAM (reliability, availability, maintainability) study, the reliability of the system and its components plays a crucial role. However the majority of unavailability comes from maintenance constraints such as crews not being available to perform a repair or lack of spare parts.
If you model an FPSO (floating, production, storage and offloading vessel) only looking at the design configuration and the reliability of the different equipment items, the production availability will likely be higher than 90% – EPCs (engineering, procurement and constructions companies) know what has worked and hasn’t worked in the past when it comes to design configuration. Even though they still have to prove their initial concept for the design, previous experiences empower the design engineer to assess some critical areas, supported by a RAM study.
In order to present a real case study, an FPSO model is shown below comprising the following system configuration:
By running the model for 250 life cycles (simulated representations of the possible life of the system), the production availability of this FPSO is 90.154% within a standard deviation of +/-0.253.
In general this is where the analyst should start measuring the maintainability of the system. Maintainability analysis includes an extensive range of factors encapsulating all the understanding related to the possible maintenance strategies. The analysis ranges from what is the size of the maintenance crew to such factors as crew mobilisation time, travel time, skills, shift patterns and so on.
For the offshore industry, this area is even more challenging. The complexity of maintenance factors can be very difficult to comprehend.
One approach would be to have all crews, spare parts, safety equipment and accessories available immediately when the repair is needed – this is represented by the example shown above where no constraints on maintenance are imposed to the model. However, this is not a realistic possibility due to constraints on space and capital investment.
A quick, more realistic example is to make one crew available to repair all the systems in the above FPSO. The result is a decrease of 2.5% on the product availability. This represents an increase of 3991 mbbls of oil in losses. Taking the oil price today, $101 bbls, the increased losses amount to $4 million dollars.
What happened? This case study includes a few hundred equipment items each with different failure modes and criticality. When only one crew is assigned to repair all the systems, the ability to account for multiple repairs disappears. From that point, the analyst will start to see a great load of maintenance activities queuing.
The maintainability analysis step of a RAM study is crucial to ensure the availability of production.