Modelling methodologies for railway asset management

Lee, Robert (2021) Modelling methodologies for railway asset management. PhD thesis, University of Nottingham.

PDF (Thesis - as examined) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Available under Licence Creative Commons Attribution.
Download (6MB) | Preview


Management of railway assets incurs significant expenditure. Railway asset management modelling can predict the cost and efficacy of an asset management plan, and thus support the asset management planning process. Modelling frameworks can be used to facilitate the development of large, multi-asset, whole life cycle models which can be used to represent large sections of rail track and associated assets. This is achieved with libraries of models and tools with a high level of inter-compatibility.

This research set out to support the development of modelling frameworks for railway asset management. It sought to determine the state of the art of railway asset management modelling in order to find which assets require further modelling development before they can be suitably represented in a framework’s model library. It also sought to determine the most accurate and suitable modelling methodology to base the framework upon.

These aims were met by first carrying out a literature review to determine the state of the art of asset management modelling for major railway asset types. This review found Petri net models solved via Monte Carlo methods to be the most suitable modelling methodology for asset management. The level crossing asset class was chosen for the development of several models to explore the different types of Petri net model, concentrating on the computational resources required. This asset class was chosen as no asset management model was found in literature, and the diversity of the asset interactions.

Literature review found several asset classes in need of further development, and some where asset management modelling may not be possible without other advances. The level crossing Petri net models developed demonstrated that computational requirements differ between the various types of Petri net. Stochastic Petri nets were found to simulate quickly, but had a high memory requirement. Coloured Petri nets were found to have the opposite requirements. A novel Petri net type, the Simple Coloured Petri net was developed to create a balance in computational cost. It was further found that complex processes such as scheduling and resource allocation can only be carried out using Coloured Petri nets due to their enhanced feature set.

This work has found that further research on modelling specific asset classes is required to enable the development of a complete asset modelling library for use in a framework. If large models are to be developed, it is recommended that the Simple Coloured Petri net be used to balance computational requirements. Any models requiring complex functions should be developed using the Coloured Petri net methodology.

Item Type: Thesis (University of Nottingham only) (PhD)
Supervisors: Andrews, John
Prescott, Darren
Keywords: Railway asset management modelling, Level crossings, Railroad crossings, Petri nets
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TF Railroad engineering and operation
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 65368
Depositing User: Lee, Robert
Date Deposited: 13 Oct 2023 07:34
Last Modified: 13 Oct 2023 07:34

Actions (Archive Staff Only)

Edit View Edit View