Digitisation of assets from the oil and gas industry: challenges and opportunities.

Moreno-Garcia, Carlos Francisco; Elyan, Eyad

Abstract

Automated processing and analysis of legacies of printed documents across the Oil & Gas industry provide a unique opportunity and at the same time pose a significant challenge. One particular example is the case of Piping and Instrumentation Diagrams (P&IDs). These are complex engineering drawings that are extensively used in the Oil & Gas industry, which contain critical information for risk assessment, and require highly skilled people to provide an accurate interpretation and analysis of their contents. This paper provides an overview of the P&IDs digitisation problem. We outline the opportunities and key challenges, discuss recent relevant work and state of the art and outline possible future direction to solve the problem. During a two-years collaborative project with an industrial partner from the Oil & Gas sector, we have encountered three main challenges other than traditional inherent image and document related challenges. These are, documents quality, skewed distribution of data and topology. In this paper, we discuss these challenges in depth and survey the main state-of-the art methodologies that may solve them.

Citation

MORENO-GARCIA, C.F. and ELYAN, E. 2019. Digitisation of assets from the oil and gas industry: challenges and opportunities. In Proceedings of 2019 International conference on document analysis and recognition workshops (ICDARW), 22-25 September 2019, Sydney, Australia. Piscataway: IEEE [online], 7, pages 2-5. Available from: https://doi.org/10.1109/ICDARW.2019.60122

• Conference Name: 2019 International conference on document analysis and recognition workshops (ICDARW)
• Conference Location: Sydney, Australia
• Start Date: Sep 22, 2019
• End Date: Sep 25, 2019
• Acceptance Date: May 15, 2019
• Online Publication Date: Sep 25, 2019
• Publication Date: Nov 7, 2019
• Deposit Date: Nov 14, 2019
• Publicly Available Date: Nov 14, 2019
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Volume: 7
• Pages: 2-5
• DOI: https://doi.org/10.1109/icdarw.2019.60122
• Keywords: Digitisation; P&ID; Engineering drawing; Machine learning; Oil and gas; Graphs
• Public URL: https://rgu-repository.worktribe.com/output/766700