““Asset Management translates an organisation’s objectives into technical and financial decisions, plans and activities” ”
Underpinning this planning and decision-making framework is good asset information that needs to be accurate, meaningful, and accessible.
Unreliable data is one of the biggest causes of inefficiency and waste. Evidence has shown that up to 30% of the total cost of ownership can be avoided by better decision-making at the design, procurement and renewal points during an asset’s lifecycle. Just collecting massive volumes of data is not enough. There’s an explosion of different data sources these days, from human-input information to data automatically collected by machines and Internet of Things (IoT)-enabled sensors and devices. As Deloitte has phrased it, businesses are now "drowning in data but starving for insights".
The challenge is ensuring Data Integrity: getting relevant, quality, reliable data that can actually be used to generate insights and inform strategic planning and effective decision making. Good asset data should be:
Accurate & reliable to ensure that decisions are made with correct information and achieving full compliance to statutory and safety obligations,
Complete & comprehensive to enable a full overview of assets to substantiate key decisions around operations, maintenance and other asset investment processes
Timely & relevant and ideally real-time, so business decisions can be made that reflect the current circumstances, particularly in volatile times
Centralised & accessible so that information isn’t lost in data silos, locked away from the specific stakeholders who need it
The risks of bad data
Poor data quality has a ripple effect across an organisation that can result in:
Large unforeseen capital expenditure occurrences to replace failed assets or worse, to repair damage to the property owned by others
Increased Work, Health and Safety (WH&S) risks
Design of assets which are not ‘fit-for-purpose’
For instance, if you don’t have access to reliable asset data, such as knowing its utilisation rates and current condition and state of repair, how can you plan for its maintenance and potential replacement? If an asset fails completely through poor maintenance, and its replacement hasn’t been scheduled, the disruption can be very expensive.
Start with good governance
Data integrity starts with robust policies governing the collection, storage and access of data, as well as its eventual deletion. Asset managers must determine what information they need and why. Collecting the specific data needed from the outset reduces ‘white-noise’ and avoids wasting limited and valuable resources.
Data policy should cover aspects such as what asset information is needed and for what purpose, how information quality can be assessed, what risk poor data presents, and how should data be managed and maintained. Ever-developing data and privacy regulations in many jurisdictions also require that attention is given to compliance.
Data repositories should be centralised, with access also centrally managed.
Automating data clean-up
According to one estimate, data scientists "spend 80% of their time cleaning and manipulating data and only 20% of their time actually analysing it".
Artificial Intelligence (AI) tools can play an important role here, scanning for data errors as well as integrating data sets while avoiding duplication. One major cause of bad data is missing values. Previously, people took "educated guesses" at this. But with machine learning, missing values can be much more accurately estimated. This may be by using a statistical approach (i.e., an average or median), or taking data from a comparable set: e.g., Building A’s lift usage or repair data may be similar for Building B, and used for predictive maintenance for Building B even if data was lost/not acquired for Building B.
It’s also vital to know when to delete data, and also when and how to archive it. Old, irrelevant data takes up space, creates confusion and in the case of certain sensitive data, must be deleted to comply with regulations. But other older data, even if no longer useful, must be retained for legal purposes such as tax records.
Data brings huge benefits to asset management. But it needs to be the right information: without Data Integrity, a digital asset management strategy won’t be successful. As asset managers migrate to digital platforms, they need to train machine learning models on high quality data, with a goal of establishing Lifecycle Data Continuity for all assets in their portfolio.
Read more about Data Continuity here:
Click/Tap to Read More