In-service inspection of static mechanical equipment: Use of a fuzzy inference system for maintaining the quality of an inspection program

It is necessary to inspect the piping components of offshore production and process facilities (OP&PFs) to investigate potential failures. This is especially vital for aging OP&PFs in order to make the necessary engineering judgments regarding maintenance and modification (M&M) activities. In an OP&PF, piping plays a vital role within the static mechanical equipment. To analyze the degradation trends in the piping, the wall thickness measurements have been periodically monitored and recorded at the locations with a high risk of failure. Inspection planners make recommendations on the thickness measurement locations (TMLs) to be monitored based on: the currently available recorded data, risk-based inspection (RBI) analysis results, plant inspection strategy guidance and other regulatory requirements. The quality of the recommendations made by an inspection planner to prioritize TMLs depends on their experience and competence. Hence, it is vital to develop expert systems to support and minimize sub-optimal decisions when an inspection planner is inexperienced. This manuscript illustrates the use of a fuzzy inference system (FIS) for making optimal in-service inspection recommendations based on the current status and trends of TMLs in the static mechanical equipment of an OP&PF. The proposed FIS enables the expertise of experienced inspection planners to be incorporated via developed membership functions (MFs) and a rule base, which will support and maintain the quality of an inspection program at the intended level.