Managing physical assets increasingly involves collecting, analysing and interpreting data. This includes degradation data, which can be combined with machine learning to predict when an asset may require maintenance.
But these analytics are only as good as the quality of the data collected. Even the best machine learning algorithms can’t compensate for inaccurate data. Until data collection and input is completely automated, there will always be the risk of human error and time lag. This affects the accuracy of data, and of its analysis.
Benefits of mobile devices for data collection
Using mobile devices is a key way to improve the accuracy and timeliness of data. For example: consider asset condition assessments undertaken on a regular basis. Ideally a maintenance technician will carry out a visual inspection, and record the results in a mobile device. This data is then uploaded to a maintenance management system.
Using mobile devices is a key way to improve the accuracy and timeliness of data.
Such an arrangement minimises the loss of data, and the inefficiencies of recording data on paper and manually entering it into a system. Data validation can also be carried out through the mobile device, to minimise the errors at the source of collection.
Despite the advances to modern maintenance management systems, most still don’t have the capability to generate accurate maintenance budgets aligned with good lifecycle management practice. Inevitably, spreadsheets or other dedicated software end up being used as well. This leads to further problems when these systems are not fully integrated. Important data such as asset age, condition and maintenance backlog will degrade over time unless data transfer is regularly carried out as part of "business as usual".
Importance of good quality data
While making data-driven decisions can be made much easier with machine learning, the currency of the data is key to supporting desired outcomes. Low quality data: whether it’s inaccurate, outdated or not relevant, has a ripple effect across an organisation.
Bad data has a ripple effect across the entire value chain of an organisation. If you capture it incorrectly, the error moves all the way through from the front end to the back end, and to compliance processes.
“Good decisions need accurate, fresh and timely data.”
This challenge is easily managed through the use of mobility solutions, the regular transfer of key asset data, and the integration of systems storing this data.