A hybrid multicriteria decision model for selecting a portfolio of risk-based maintenance actions in natural gas pipelines

• Published in: Journal of natural gas science and engineering Vol. 103; p. 104655
• Main Authors: Viana, Francisco Filipe Cunha Lima, Casado, Ramon Swell Gomes Rodrigues, da Silva, Lucas Borges Leal, Alencar, Marcelo Hazin, Ferreira, Rodrigo José Pires, de Almeida, Adiel Teixeira
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.07.2022
• Subjects: Maintenance planning; Multicriteria portfolio evaluation; Multidimensional risk assessment; Natural gas pipelines; Multicriteria portfolio evaluation; Multidimensional risk assessment; Natural gas pipelines; Maintenance planning
• ISSN: 1875-5100
• DOI: 10.1016/j.jngse.2022.104655

Accidents in natural gas pipelines are a significant concern because of the potential catastrophic consequences for humans, the environment, and properties. To tackle this matter, risk-based maintenance approaches are used for decision-making to prioritize appropriate mitigation actions. However, selecting maintenance actions faces other organizational limitations that preclude companies from managing pipelines more effectively. This paper proposes a decision model for selecting a portfolio of risk-based maintenance actions in natural gas pipelines under environmental, financial, and human considerations. The model uses the multicriteria c-optimal PROMETHEE method to select maintenance actions that best meet organizational objectives. Moreover, the model also adopts criteria for categorizing risks to pipeline sections based on a multidimensional perspective. A numerical application is applied to validate the proposed model, considering a natural gas pipeline that is divided into eight sections. For this pipeline, 40 maintenance actions are evaluated according to the risk, complexity and maintenance time involved. Relevant portfolios are compared and an extensive sensitivity analysis is conducted. Results show that by using c-optimal PROMETHE, three optimal portfolios are obtained. Of these portfolios, 29 actions are selected with the best performance in the criteria, and assuming that maintenance departments have an effective budgetary allocation. Therefore, this proposed approach gives better support to the recommendations made to the decision-maker. •A multicriteria model is presented to select portfolio of maintenance actions for NGP.•Maintenance actions are evaluated according to risk, complexity and maintenance time.•The proposed methodology support decisions on minimizing of multidimensional risks.•A numerical application presented robust results in obtaining the best portfolio.