An open modelling approach for availability and reliability of systems. (March 2019)

• Record Type: Journal Article
• Authors: Penttinen, Jussi-Pekka; Niemi, Arto; Gutleber, Johannes; Koskinen, Kari T.; Coatanéa, Eric; Laitinen, Jouko
• Is Part Of: Reliability engineering & system safety. Volume 183(2019)
• Page Start: 387
• Page End: 399
• Publication Date: 2019-03
• Subjects: Risk assessment technique -- System performance -- Complex systems -- Dynamic modelling -- Fault tree analysis -- Model data format; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2018.11.026
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Object-oriented approach for risk and performance assessment of complex systems. Customized domain-specific features can be included in standard analysis techniques. Example combines fault tree, dynamic process phase, and production function models. Open data format to define modelling technique, model structure and parameter values. Applicable to industrial process reliability, availability and performance analyses. Graphical abstract: Abstract: This paper introduces an Open Modelling approach for Availability and Reliability of Systems (OpenMARS), which is developed for risk and performance assessment of large and complex systems with dynamic behaviours. The approach allows for combining the most common risk assessment and operation modelling techniques. This ensures a high degree of freedom for the modeller to accurately describe the system without limitations imposed by an individual technique. OpenMARS uses a platform-independent tabular format to define the used modelling technique, to create the model structure, and to assign the parameter values. We developed the format to enable a straightforward manual model definition while maintaining database compatibility. This paper also presents our calculation engine for stochastic simulation-based analysis of OpenMARS models. Our intention is to use this approach as a basis for new software. We demonstrate the feasibility of OpenMARS with an example of a multi-state production process that is subject to failures. The example creates a comprehensive system model by combining interconnected failure logic, operation phase, and production function models. We believe that the advanced features of OpenMARS have wide ranging applications for analysis of reliability, performance, and energy efficiency of complex industrial processes.