The Siri for Maintenance program aims to unlock the valuable information captured in unstructured documentation collated by maintainers and engineers. Such documentation includes maintenance work order records, failure modes and effects analysis, maintenance procedures, original equipment manufacturer’s documentation, root cause analysis, and safety investigations. We want to make relevant information from these records machine readable. In other words, a machine can understand the semantic intent of the text and use this to support inference and reasoning without human support.
The Siri for Maintenance team at UWA includes Dr. Wei Liu and Dr. Tim French, Melinda Hodkiewicz, and PhD students Caitlin Woods and Tom Smoker. We have established a number of important overseas collaborations. These include, but are not limited to, the following.
The Systems Integration Team at the US Government’s National Institute of Standards and Technology (NIST) Laboratory in Maryland USA to work on natural language processing and application of human factor approaches to maintenance data collection and error management. In 2019 took part in NIST’s Standards Requirements Workshop for Natural Language Analysis workshop and have been involved in supporting the NIST team’s development of the Nestor tagging tool.
The Sirius Laboratory for Scalable Data Access at the University of Oslo, funded by the Norwegian Oil and Gas industry. We are working with experienced ontologists on ontology development and the use of templates to efficiently ingest data from SAP and other data sources into ontologies for reasoning.
Playing an active role in the Industrial Ontology Foundry (IOF), an international group of ontologists aimed at developing a reference ontology for manufacturing. This includes leading the Maintenance Working group, participating on various committees and running their web site. In February 2019 the UWA team coorganised the International Industrial Ontologies workshop in Oslo. This was attended by 140 delegates from 43 companies, 24 research institutions and 18 countries.
Collaboration with IOF colleagues in the US and France to produce an open, sharable ontology for maintenance aligned to an upper ontology (BFO). BFO is now an ISO Standard.
The Siri for Maintenance project received support from the Datacentric Engineering group at the Alan Turing Institute (ATI) in London in 2018. This support allowed Professor Melinda Hodkiewicz to spend time at the Turing and collaborate with a number of UK academics and industry players.
In 2019 our work was given a major boost due to the establishment of a 5 year $4M Australian Research Council grant, matched by similar funding from three industry players (BHP, Alcoa and Roy Hill), to establish a Centre for Transforming Maintenance through Data Science. There are three research themes, one called “Supporting the Maintainer” which is about the application of NLP and ontology tools to maintenance and similar records. All of the academics associated with the Siri for Maintenance program are involved.
Outputs from the Siri for Maintenance project collaborations:
Martin G. Skjæveland, Johan W. Klüwer, Melinda Hodkiewicz, Leif Harald Karlsen and Daniel P. Lupp, Scalable construction of sustainable knowledge bases, Tutorial at ISWC 2019, Auckland NZ, 26-30 October 2019.
Brundage, M. P., Sexton, T., Hodkiewicz, M., Morris, K., Arinez, J., Ameri, F., Ni, J., and Xiao, G. (July 22, 2019). “Where Do We Start? Guidance for Technology Implementation in Maintenance Management for Manufacturing.” ASME. J. Manuf. Sci. Eng. September 2019; 141(9).
Karray, M.H., Ameri, F., Hodkiewicz, M. and Louge, T., 2019. ROMAIN: Towards a BFO compliant reference ontology for industrial maintenance. Applied Ontology, (Preprint), pp.1-24.
Hastings, E., Sexton T, Brundage MP, Hodkiewicz M., 2019. Agreement Behavior of Isolated Annotators for Maintenance Work-Order Data Mining. In PHM Society Conference, Phoenix, Sep 23-26 2018.
Sexton T, Hodkiewicz M, Brundage MP., 2019. Categorization Errors for Data Entry in Maintenance Work-Orders. In PHM Society Conference, Phoenix, Sep 23-26 2018.
Sexton T, Hodkiewicz M, Brundage MP, Smoker T. Benchmarking for Keyword Extraction Methodologies in Maintenance Work Orders, 2018. In PHM Society Conference, Philadelphia, Sep 24-27 2018.
Smoker, T., French, T., Liu, W., and Hodkiewicz, M., 2017. Applying cognitive computing to maintainer-collected data, ICSRS 2nd International Conference on System Reliability and Safety, December 20-22, 2017 Milan, Italy.
And previous relevant outputs.
Hodkiewicz, M.R., Batsioudis, Z., Radomiljac, T., Ho, M.T.W., 2017. Why autonomous assets are good for reliability – the impact of ‘operator-related component’ failures on heavy mobile equipment reliability, Annual Conference of the Prognostics Health Management Society, 2-5 October, St. Petersburg, Florida
Ho, M.T.W., 2016. A shared reliability database for mobile mining equipment, Ph.D. thesis, University of Western Australia.
Hodkiewicz, M. and Ho, M.T.W., 2016. Cleaning historical maintenance work order data for reliability analysis. Journal of Quality in Maintenance Engineering, 22(2), pp.146-163.
M Ho, MR Hodkiewicz, CF Pun, J Petchey, Z Li, 2015. ‘Asset Data Quality—A Case Study on Mobile Mining Assets’, Engineering Asset Management-Systems, Professional Practices and Certification. Springer International Publishing. 335-349.
One of our greatest assets is our close links to the maintenance and asset practitioners here in the resources and infrastructure community in Perth WA. Perth is a globally significant location for mining and oil and gas with BHP, Rio Tinto, Shell, Chevron and other global players having significant investments and staff located here.
We are keen to talk to prospective PhD students with an interest in applied ontology and natural language processing.
We also offer PhD, Honours and Masters of Professional Engineering and Master of Data Science projects.