Coiled tubing has been an invaluable tool to the oil and gas industry for some forty years. Each time that coiled tubing is rigged up and run into a well it carries with it a risk that the objective of that run will not be met. This paper analyses that risk based on some two years continuous records in the Asia Pacific Region and finds that the average rate of a successful run is 82% over a 23-month period. A database of about 1200 runs demonstrates the type, cause and frequency distribution of failures. Risk management options are reviewed and potential solutions offered.

The cause of failures is most frequently attributed to unknown well conditions. However, inadequate job planning, operator errors and equipment failures also have a significant contribution. The levels of technical complexity demands are escalating. In balance, there are increasing operational skill levels, equipment reliability improvements and a better focus on pre-job planning. This, then, tends to hold the status quo at a constant frequency over time. In order to increase the rate of success, lowering the difficulty of the jobs attempted and/or increasing the operating and equipment reliability will be required. Simplicity and reliability tend to go hand in hand. However, there is a general industry temptation to over-engineer and overcomplicate both the work scope and the equipment. The result may be viewed as unreliable and perhaps, not cost effective coiled tubing solutions.

The opportunity exists to significantly reduce the run failure count. This could be achieved by lowering ambitions and allowing equipment and skill levels to ‘catch up’ with established industry coiled tubing solutions. It then raises the question "Is it better to do something simple well or to risk failure with technical ambition?" This paper concludes that the current failure rate can be improved upon with various corrective actions. This is essential in order to progress to further technical advancements.

You can access this article if you purchase or spend a download.