This dataset contains both raw (columns 1-5) and cleaned data derived from 5486 maintenance records (Work Orders) pertaining to 5 similarly sized excavator/ shovels at a variety of different minesites across Australia over 10 years. The data was initially used to calculate the reliability of components necessary for the comfort and safety of the operators. A second file with only the raw data is also supplied as part of the download so the user can develop or utilise alternative cleansing algorithms if desired. The cleaning work was done prior to the availability of NLP and uses a rule-based lexical approach and expert knowledge.

Details on the data and how it was collected are available in the open source publication available at http://papers.phmsociety.org/index.php/phmconf/article/view/2449

Required data set citation is https://prognosticsdl.systemhealthlab.com/ and the date you accessed the data.

Please cite this publication if you use this data. Hodkiewicz, M.R., Batsioudis, Z., Radomiljac, T. and Ho, M.T., 2017. Why autonomous assets are good for reliability–the impact of ‘operator-related component’failures on heavy mobile equipment reliability. In Annual Conference of the PHM Society (Vol. 9, No. 1).

Note PM01 work type is corrective work, PM02 is preventative work (work based on maintenance strategy) and $ are in Australian $. Data is from 2002-2012.

If you are having issues downloading the data, please reach out to the author by email at the address below.