A Multi-attribute Quantitative Assessment System for Offshore Facilities’ Decommissioning Options Decision-making

Abstract

Since offshore oil and gas resources have been developed widely worldwide for decades, some facilities have reached the end of their service lives and need to be decommissioned. However, due to the constraints of relevant regulations, technologies, costs, potential risks, and environmental impacts, what options should be taken for decommissioning offshore facilities and whether the chosen method can be convincing have always been puzzling the offshore industry. This research is devoted to developing a multi-attribute quantitative evaluation system called MADM-Q (Multi-attribute Decision Making-Quantitative), which can quickly provide decision makers with reliable quantitative evaluations in the engineering planning stage and choose the most reasonable decommissioning options. The proposed system refers to the current comparative assessment method in the UK and sets up three assessment sub-modules: cost, risk, and impact. The first sub-model, Engineering Cost Evaluation System (ECES) is the first bottom-up approaches cost assessment model in offshore facility decommissioning research. This project uses 26 reports, including 32 facilities’ historical data of decommissioned platforms in the North Sea to establish two cost assessment models and compares the model performance. The comparison results show that the ECES used more data categories (7 categories minimum) than top-down approach (5 categories minimum) and has better accuracy than top-down framework model built by other researchers. The second sub-model, innovative Hierarchical Analyst Domino Evaluation System (HADES) acts as a risk assessment module, considering Domino Effect Accidents (DEAs) and providing accurate and rapid quantitative risk assessment results. The HADES is a quasi-dynamic quantitative risk assessment system constructed by combining the core idea of AHP with traditional QRA. It defines the trigger mechanism of DEAs in two sufficient and necessary principles and builds a new accident scene generation mechanism based on the two principles. This method is different from the scenario hypothesis method used by traditional QRA but allows the system generate accident scenarios dynamically according to the accident situation and build the DEAs causality network. The third sub-model, Composite Impact Evaluation System (CIES) assesses the environmental and socioeconomic impacts of the project as one of the cost modifiers. This part mainly considers the marine environmental and social impact caused by the hydrocarbon leakage accident and converts the impact results into monetary form as part of the project reserve cost that energy companies need to prepare. In addition, there are also data exchanging among the three modules: • The ECES results will serve as the basis. • The HADES results will be incorporated to remind decision-makers to prepare for reserve cost.

• The HADES results will be fed into the CIES to quantify the project’s negative impacts, which may increase reserve cost. According to the requirements of decision makers for different decommissioning options, the system can quickly obtain the basic cost range and the Individual Risk Per Annum (IRPA) value of the project by only using the basic information of the offshore facility itself and the location information as input. The three modules in this system have a mature framework, and the detailed methods used are based on physics and general-purpose industrial equations. It can be applied to different regions of the world only by regional adaptation of commercial data. Among them, after comparing the evaluation results of ECES and HADES with the actual results of the case report, it reflects the excellent accuracy of the ECES and HADES systems - 12% cost evaluation average deviation and 1.43E-04 IRPA Evaluation bias. In addition, the three modules provide many access ports, allowing users to use more optimized methods, such as the results obtained by finite element analysis as input values to obtain more accurate accident scene development models and evaluation results. Among them, although the HADES method is currently only a quasi-dynamic risk assessment method, it can be developed into a real-time dynamic risk monitoring and assessment system by combining advanced technologies such as digital twin technology and sensor technology, which has excellent development potential.