Confidently extracting hierarchical taxonomy information from unstructured maintenance records of industrial equipment

Abstract

Maintenance records of complex industrial equipment contain a large amount of unstructured data (e.g. technician notes) pertaining to repair actions and associated equipment sub-components, degradation conditions, failure mechanisms, etc. These unstructured data can yield valuable insights to improve the equipment design and maintenance plans, resulting in higher productivity and lower operating costs. Since manual review of information is time-consuming, companies make limited use of the maintenance records. To address this opportunity, we propose a taxonomy-guided method for automatically analysing the unstructured data and inferring critical information, specifically the hierarchy of the equipment's sub-assemblies and constituent parts that malfunctioned or failed during a breakdown event. Our method leverages syntactic (related to word frequency) as well as semantic (related to word co-occurrence and their meaning) knowledge. A novel contribution of our work is that we provide a confidence score for the information inferred by our method. Only the maintenance records which receive a low confidence score will require manual review to confirm the automated method's results, thus ensuring minimal use of human resources. We demonstrate the performance of our method using a real-world data set from equipment used in oil rigs.

Keywords:

• Taxonomy
• ontologies
• product life cycle
• maintenance records
• non-parametric
• confidence score
• unsupervised learning

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

In this paper, we have illustrated the application of our methodology using real-world data from an industrial partner. This data cannot be made available as it is proprietary to the company.

Additional information

Notes on contributors

Abhijeet S. Bhardwaj

Abhijeet S. Bhardwaj received his Integrated M.Tech. degree in Geophysical Technology from the Indian Institute of Technology Roorkee, Roorkee, India, in 2017. He is working towards the Ph.D. degree at the Department of Industrial and Systems Engineering, University of Wisconsin–Madison, Madison, WI, USA. His research interests include unstructured natural language data mining and analytics, industrial prognosis using data fusion and decision making for Internet-of-Things-enabled smart and connected systems.

Dharmaraj Veeramani

Dharmaraj Veeramani received the B.S. degree in mechanical engineering from the Indian Institute of Technology Madras, Chennai, India, in 1985, and the M.S. and Ph.D. degrees in industrial engineering from Purdue University, West Lafayette, IN, USA, in 1987 and 1991, respectively. He is the E-Business Chair Professor with the Department of Industrial and Systems Engineering, University of Wisconsin-Madison, Madison, WI, USA. His research focuses on emerging frontiers of digital business, Internet of Things technologies and applications, smart and connected systems, and supply chain management.

Shiyu Zhou

Shiyu Zhou received the B.S. and M.S. degrees in mechanical engineering from the University of Science and Technology of China, Hefei, China, in 1993 and 1996, respectively, and the master’s degree in industrial engineering and the Ph.D. degree in mechanical engineering from the University of Michigan, Ann Arbor, MI, USA, both in 2000. He is the Vilas Distinguished Achievement Professor with the Department of Industrial and Systems Engineering, University of Wisconsin-Madison, Madison, WI, USA. His research interests include data-driven modeling, monitoring, diagnosis, and prognosis for engineering systems with particular emphasis on manufacturing and after-sales service systems.