Prediction and Evaluation of Accidents within Oilfields of Arabian Gulf Company Using Discriminant Analysis
Article Sidebar
Main Article Content
Abstract
The dynamic nature of oil and gas production is one of the major causes for various types of accidents resulting in injuries and fatalities in oil fields. The main purpose of this paper is to identify and classify accidents in Arabian Gulf Oil Company (AGOCO) through application of the Linear Discriminant Analysis (LDA) technique. Data collected were for 8 years spanning from 2005 to 2012, from four oil fields (Sarir, Nafoora, Messla, and Byda).
The LDA is used to classify an accident into one of the accident groups; “Oil and gas Leak”, “Fire”, “Accident”, and “Damage”.
The developed discriminant functions revealed significant association between groups 60.1%, 22.75%, and 6.97% of between groups variability. However, the structure matrix revealed two significant predictors only of the first function, namely production area and camp with scores of 0.791 and 0.766 respectively. For the second function revealed also two significant predictions; the oil well and the transition station with scores of 0.806 and 0.740 respectively.
The third function has no significant predictors. The cross-validated method showed that 65.1% of the grouped cases are correctly classified.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
J. I. Chang, and C. C. Lin, ―A study of storage tank accidents,‖ Journal of Loss Prevention in the Process Industries, vol. 19, pp 51–59, 2006. DOI: https://doi.org/10.1016/j.jlp.2005.05.015
W. Wei, and L. Mao, ―Analysis of accident in production in oilfield of China,‖ Center for Urban Public Safety, Nankai University, Tianjin 300071, China, 2003.
T. Hastie, R. Tibshirani, and J. Friedman, ―The elements of statistical learning - data mining, inference, and prediction,‖ Springer Series in Statistics, New York, 2001. DOI: https://doi.org/10.1007/978-0-387-21606-5
B. Efron, and R. Tibshirani, ―Cross-validation and the bootstrap: estimating the error rate of a prediction rule,‖ Technical Report, University of Stanford, 1995.
Multivariate Data Analysis Using SPSS Discriminant Analysis: Lesson2. (2008), Available: http://www.researchgate.net