QUANTITATIVE RISK ASSESSMENT OF SUBSEA PIPELINE AT NATUNA SEA

• Main Author: Alvin Tanujaya, Vincent
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/66723
• Institution: Institut Teknologi Bandung

The Natuna Sea is the outermost marine territory of Indonesia which is traversed by one of the most important international shipping lanes in the world, The Suez Canal Route, which connects Asia, Africa, and Europe. It connects the two largest and busiest container ports in the world, the Port of Shanghai and the Port of Singapore. In addition, at least one-third of the world's demand for oil and natural gas are distributed through this route. Therefore, the area of the Natuna Sea is generally traversed by large vessels. The density of marine traffic in the Natuna Sea is a potential risk for offshore oil and gas facilities which operate in that area, and the subsea pipeline is one of the facilities to consider. According to the DNVGL RP-F107, there are three major risks to analyze: dropped anchor, dragged anchor, and vessel sinking. Therefore, this study is consists of risk assessment and analysis of subsea pipelines based on vessel traffic database or Automatic Identification System (AIS) data. In this study, AIS data will be processed and analyzed to determine the probability of ship distribution on the main shipping lane for the entire data coverage area. Thus, it can be used to assess the risk of subsea pipelines throughout the coverage area. The probability of ship distribution in this study will then be calculated using three methods: actual frequency data, a fitting test of multi-peak Gaussian distribution, and Monte Carlo probabilistic simulation. Based on the AIS data processing results, it indicates that the majority of vessels are known to operate on international shipping lane, that is in the Northeast (15° to 45°) and Southwest (195° to 225°) directions with a probability of 77.52% and shows a relatively constant annual trend. Therefore, this study is then more focused to analyze the effect of marine traffic in the dominant direction on the potential risk of subsea pipelines. The calculation method of the probability of distribution of ship data using the fitting test of 5-peak Gaussian distribution shows a more accurate result than the Monte Carlo simulation, with an error of 0.01% and a coefficient of determination of 0.9995. After obtaining the 5-peak Gaussian distribution equation, the research can be continued to analyze the consequences and the probability of failure of the subsea pipeline. In the analysis of the structural consequences, ship data are classified into minor damage, moderate damage, major damage, and rupture based on the ratio between local deformation and total pipe diameter. Then, for each class of structural consequences, classification is carried out based on the potential for pipe leakage into no release, leakage, and rupture. Meanwhile, an analysis of the probability of failure is carried out for each consequence class by taking into account the probability of ship distribution, the probability of occurrence of external activities, and reduction factors. The reduction factors used in this study are the application of VTS (Vessel Traffic System), stand-by vessel, and pipeline registration to the navigational chart. The results of the risk assessment and analysis of the reviewed subsea pipeline are mostly at an acceptable risk level, except for the risk of rupture scenario due to dropped and dragged anchor which is at the ALARP level. Finally, risk assessment and analysis are conducted on the other two comparison pipelines to analyze the relevance of the application of the 5-peak Gaussian distribution to determine the potential failure risk of any pipeline around the main shipping lane. Based on the results of the risk comparison analysis between the reviewed pipeline and comparison pipelines, it can be concluded that the farther the position of the pipe from the center of the shipping lane and the greater the angle between the pipe and the shipping lane, the frequency of ships can be significantly reduced and the dominating direction category can also change. Therefore, the 5-peak Gaussian distribution is only relevant to determine the distribution of ships on the main shipping lanes in the coverage area. Thus, to determine the probability of ship traffic on a pipeline which is relatively far from the center of the shipping lane, it must be analyzed by considering the local distribution in the pipeline area.