Condition Prediction Models for Oil and Gas Pipelines Using Regression Analysis Mohammed S. El-Abbasy 1 ; Ahmed Senouci 2 ; Tarek Zayed, M.ASCE 3 ; Farid Mirahadi 4 ; and Laya Parvizsedghy 5 Abstract: Although they are the safest means of transporting oil and gas products, pipelines can sometimes fail with hazardous conse- quences and large business losses. The decision to replace, repair, or rehabilitate depends mainly on the condition of the pipeline. Assessing and predicting its condition is therefore a key step in the maintenance plan of a pipeline. Several models have recently been developed to predict pipeline failures and conditions. However, most of these models were limited to the use of corrosion as the sole factor to assess the condition of pipelines. The objective of this paper is to develop models that assess and predict the condition of oil and gas pipelines based on several factors including corrosion. The regression analysis technique was used to develop the condition prediction models based on historical inspection data of three existing pipelines in Qatar. In addition, a condition assessment scale for pipelines was built based on expert opinion. The models were able to satisfactorily predict pipeline condition with an average percent validity above 96% when applied to the validation data set. The models are expected to help decision makers assess and predict the condition of existing oil and gas pipelines and hence prioritize their inspection and rehabilitation planning. DOI: 10.1061/(ASCE)CO.1943-7862.0000838. © 2014 American Society of Civil Engineers. Author keywords: Oil and gas pipelines; Condition prediction; Regression analysis; Quantitative methods. Introduction Onshore and offshore pipelines are the most commonly used means of transporting crude oil and gas in the world nowadays because of their safety and economy. Unfortunately, the number of accidents due to aging pipelines in operation has significantly increased (Teixeira et al. 2008). Aging pipelines can be affected by a range of corrosion mechanisms, which may lead to a reduction of their structural strength and eventual failure (Ahammed 1998; Netto et al. 2005; Teixeira et al. 2008). Corrosion is an important cause of pipe- line deterioration due to aggressive environments (Ahammed and Melchers 1996). Despite that, corrosion is not the only factor responsible for pipeline deterioration. Other factors, such as third- party activity, incorrect operation of pipelines, inadequate design, faulty construction, outside forces, and many others, can have a negative impact on pipeline condition. The majority of developed condition assessments or failure pre- diction models are not considered comprehensive because they deal with only one failure cause such as corrosion. In addition, no stan- dard condition assessment scale for oil and gas pipelines (condition rating system and its associated rehabilitation actions) has been developed yet. Therefore, there is a need for the development of a more comprehensive condition assessment model and a standard condition assessment scale for oil and gas pipelines. This research is intended to include other criteria besides cor- rosion in the pipeline condition assessment process and to build a condition assessment scale framework. In order to address these issues, the presented research thoroughly identified the main crite- ria that affect the condition of oil and gas pipelines. The influential criteria are selected based on expert opinions and from the litera- ture. Finally, a model to assess the condition of oil and gas pipelines is developed and validated. The model will help pipeline operators to assess the condition of oil and gas pipelines and can be used as a guideline for future assessment prediction. Research Objectives The main objectives of the present study are to 1. Identify and study the critical factors affecting the condition of oil and gas pipelines, 2. Build a condition assessment scale, 3. Design condition assessment and prediction models, and 4. Develop expected condition deterioration curves for such pipelines with respect to age. Background Previous Studies Major efforts have been carried out in the last decade to assess the condition of oil and gas pipelines. Noor et al. (2010, 2011) used semiprobabilistic and deterministic methodologies to predict the remaining strength of submarine pipelines subjected to internal 1 Ph.D. Candidate, Dept. of Building, Civil and Environmental Engi- neering, Concordia Univ., Montreal, QC, Canada H3G 1M8 (correspond- ing author). E-mail: msksia@yahoo.com 2 Associate Professor, Dept. of Civil and Environmental Engineering, Qatar Univ., P.O. Box 2713, Doha, Qatar. E-mail: a.senouci@qu.edu.qa 3 Professor, Dept. of Building, Civil and Environmental Engineering, Concordia Univ., Montreal, QC, Canada H3G 1M8. E-mail: zayed@ encs.concordia.ca 4 Graduate Student, Dept. of Building, Civil and Environmental Engineering, Concordia Univ., Montreal, QC, Canada H3G 1M8. E-mail: faridmirahadi@gmail.com 5 Ph.D. Candidate, Dept. of Building, Civil and Environmental Engineering, Concordia Univ., Montreal, QC, Canada H3G 1M8. E-mail: l.sedghy@yahoo.com Note. This manuscript was submitted on August 2, 2013; approved on December 31, 2013; published online on March 6, 2014. Discussion period open until August 6, 2014; separate discussions must be submitted for indi- vidual papers. This paper is part of the Journal of Construction Engineering and Management, © ASCE, ISSN 0733-9364/04014013(17)/$25.00. © ASCE 04014013-1 J. Constr. Eng. Manage.