Articles

« go back

Maintenance inspection design software for full-scale industrial systems

Asset inspection and maintenance is an important activity. The ability to use data from inspections wisely, and to select how and when to inspect next, offers the opportunity for improved decision making: better use of inspection and maintenance resources, and reduced risk of asset failure. At present, statistical tools for inspection analysis and planning are usually based on separate and independent analyses of individual components.

Compared to existing methods for inspection analysis and planning, the model introduced here is novel in that:

  • It allows direct analysis of large multivariate systems, rather than modelling components separately and independently
  • It permits use of data from partial inspections at arbitrary times, yet we learn about the whole system
  • We learn about the uncertainties in wall thickness and corrosion rate per component, and the dependencies between these across the system, rather than assuming these uncertainties and dependencies known and fixed beforehand
  • It allows economically-optimal future inspection strategies to be estimated consistently.

 

In due course, the intention is to implement the method, with accompanying algorithms for optimal inspection design, within Shell's S-IDAP software.

 

Project details
The proposed work plan would involve the development of prototype Bayesian Inspection Planning software.

Software would be in the form of a MATLAB toolbox (for Windows / Linux), with major challenges including development of a useable user interface, particularly for problem specification this might be a set of input files expected by the software initially, evolving to GUIs. Issues of elicitation of expert beliefs will also provide a challenge to be incorporated into any such software.

Efficient implementation of existing prototype algorithms would be needed for methods to work as production code. The generic nature of methodology would make it applicable widely. However there will be an initial application to inspection planning allowing software to be used as a decision support tool.

 

Project staff and support
David Randell (Intern, University of Durham)
Philip Jonathan (Company Supervisor, Shell Projects and Technology)
Michael Goldstein (Academic Mentor, University of Durham)
Lorcan Mac Manus (Technology Translator, Industrial Mathematics KTN)

This Internship project is being carried out at the Shell Projects and Technology, in conjunction with the University of Durham. It is part of the KTN's Industrial Mathematics Internships Programme, co-funded by EPSRC. Start date: April 2010; duration: 6 months.

 

Related resources:
Other energy and utilities projects
Other materials projects
Other Internship projects

Comments

Comments

6 people have had something to say so far

Related this this work, the article below has been accepted to the Journal of Risk and Reliability.

https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=70922&­folderId=1136176&name=DLFE-11174.pdf

The abstract is as follows:
Modelling of complex corroding industrial systems is critical to e ffective inspection and maintenance for assurance of system integrity. Wall thickness and corrosion rate are modelled for multiple dependent corroding components, given observations of minimum wall thickness per component. At each inspection, partial observations of the system are considered. A Bayes Linear approach is adopted simplifying parameter estimation and avoiding often unrealistic distributional assumptions. Key system variances are modelled, making exchangeability assumptions to facilitate analysis for sparse inspection time-series. A utility based criterion is used to assess quality of inspection design and aid decision making. The model is applied to inspection data from pipework networks on a full-scale o shore platform.
Posted on 21/07/10 15:08.
Furthermore, there is a presentation given to Lancashire and East Cumbria local group of the Royal Statistical Society available at the URL below:

https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=70922&folder­Id=1136176&name=DLFE-11175.pdf
Posted on 21/07/10 15:55 in reply to Lorcan Mac Manus.
The paper Bayes linear learning about large industrial systems has been accepted to RSS 2010. The abstract for the paper is given below:

Bayes linear learning about large industrial systems

Objectives

Careful inspection and maintenance of large industrial systems is essential to assure system efficiency and integrity. For example manufacturing systems, susceptible to corrosion, require careful monitoring and maintenance to avoid economic and environmental costs. Modelling these complex systems can improve the effectiveness of inspection and maintenance activities. Typically, system components are modelled independently despite their inter-dependence, to make analysis tractable. Here, we adopt a Bayes linear framework to facilitate multivariate modelling of the full system. Model forecasts provide a basis for future inspection design.

Methods/Models

A Bayes Linear approach is adopted, simplifying parameter estimation and avoiding often-unrealistic distributional assumptions. Exchangeability assumptions are made to facilitate analysis of sparse inspection time-series. We estimate system evolution in time, together with key model variances. As a result, future system measurement campaigns can be designed with respect to an utility-based expected loss criterion, to aid decision making.

Results and Conclusions

Evolution of wall thickness and corrosion rate and their dependence for multiple corroding components in a large industrial system are modelled. Exploiting dependence structure improves forecast quality, allowing optimal design of future inspection campaigns.
Posted on 21/07/10 15:57 in reply to Lorcan Mac Manus.
Also arising from this work, please see the link below for the slides from a talk given at the Royal Statistical Society Conference in Brighton Sept 2010.

https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=70922&folderI­d=1136176&name=DLFE-16867.pdf
Posted on 07/10/10 11:56.
The final version of the case study for this project is now available at:

https://ktn.innovateuk.org/c/document_library/get_file?p_l_id=70922&folderId=­1027484&name=DLFE-16873.pdf
Posted on 07/10/10 16:15.
David presented his work at the Alan Tayler Day this year. The presentation is available at the link below:

https://ktn.innovateuk.org/c/document_library/get_file?uuid=9cae96ba-383f-­4034-9405-daff7534eb54&groupId=47465
Posted on 08/12/10 10:27.

Most read articles

Maintenance inspection design software for full-scale industrial systems

Asset inspection and maintenance is an important activity. The ability to use data from...

Special Interest Group in Environmental Risk Management

The Industrial Mathematics Special Interest Group (SIG) in Environmental Risk Management was...

Green buildings-thermal comfort and energy consumption

The objective of this Industrial CASE PhD project is to develop novel mathematical models for...