Intelligent maintenance
At this year's AMPEAK 2023 conference, AssetFutures’ Darren Chuang tackled the topic of intelligent maintenance and presented a data-driven approach to it: digitally calibrated replacement decisions. This approach empowers asset managers to efficiently handle large portfolios and make confident, strategic decisions using technology.
Asset management often feels like balancing on a tightrope. Managers must ensure a high level of service but have to plan and operate within limited funding constraints. Downtime must be minimised, however replacing assets too early can be wasteful.
Thankfully, advancements in technology and the use of automatic calibrating algorithms are enabling smarter and more accurate decision-making. Rather than relying on intuition, digital systems can now analyse asset data and provide answers to questions like,"How can I be certain that replacing an asset now is a better option than replacing it later?" This empowers asset managers to make decisions with confidence.
⚖️ Balancing service vs cost
Intelligent maintenance helps organisations achieve the optimum balance of service and cost. But achieving this requires a commitment to data collection, governance, and analytics.
Organisations must have effective control and governance over their assets to comply with the ISO 55000 standard for Asset Management. This includes managing risk and opportunity to achieve the desired balance of cost, risk, and performance, ultimately realising value.
All assets deteriorate over time, and predicting how they degrade can have a significant impact on asset management outcomes. At AssetFuture, we have a core library of degradation models to generate plans that make these predictions. Each asset passes through three stages as it degrades, from potential failure (low possibility) to functional failure (high possibility).
Deciding when the appropriate intervention point depends on what maintenance approach is adopted: Reactive, Preventative, or Intelligent (Predictive or Prescriptive).
Reactive maintenance is firefighting: fixing problems after they occur. It has the lowest cost but the highest number of failures and unplanned downtime.
Preventative maintenance repairs assets before they fail but has the highest cost and waste in intervening early.
Intelligent maintenance balances the previous two approaches by using data to determine the optimal time to intervene. The challenge is the time and commitment required to achieve data capture and database maintenance.
Balancing act of maintenance
It’s a balancing act between trade-offs to get to an optimum point, timing accuracy is key.
🤔 Analysing roof replacement data
Many government agencies have been investing in the level of data management required to enable intelligent maintenance. Having accrued actual evidence from 2017 onwards, they can now quantitatively demonstrate specific, real degradation and intervention points within the built asset register.
We conducted a study comparing theoretical and actual intervention to determine model calibration and generate more accurate insights. We analysed historical roof replacement data from 2020 to 2022, plotting the roof replacement count by intervention condition. Roof sheeting, which is the largest portfolio element value, represents around 16% of the total value.
We then executed data analysis to determine the most effective way to group the data, partitioning it by material types (aluminium sheet, concrete tile, steel sheet etc) and climate zones (beachfront, coastal, arid etc). By factoring in climate zones we could more accurately predict the degradation rate for different materials in different locations. This enables more informed decisions about when to intervene and how to optimise asset management outcomes.
Overall, we found that roofing materials are lasting longer than anticipated, as replacements were occurring much later than expected. Steel sheet roof lasted 5.19 years longer than predicted. This discrepancy highlights the crucial importance of model calibration, ensuring that outcomes more accurately reflect real-life situations.
✨ Calibration and data quality
For intelligent maintenance to be effective, calibrating the Intervention Condition needs to be an ongoing process that requires regular review. Each calibration makes the degradation model more accurate over time. To further increase the accuracy of the results, it's crucial to improve the quality of the source data collected.
At AssetFuture, we have a strong track record of helping organisations with data collection, governance, and analytics. We know that a lack of reliable, high-quality data is a continual challenge for Asset Management, but through platforms like AssetFuture, intelligence maintenance is becoming an achievable reality.
As a part of our commitment to #empowertomorrow, we invite you to connect with us and start your journey towards maintaining intelligence and strengthening your portfolio alongside your team of operators. Contact us for more information and take the first step towards achieving success.