Fast-tracking to Data Maturity

Lifecycle modelling has big benefits for Asset Management. It enables decision-making for operational maintenance and replacement of assets. It also allows far more accurate budgeting and certainty around cost. If we have a greater understanding of assets, we can more accurately predict their needs and avoid critical failure. And we can manage them in a much more efficient way, for risk mitigation and cost and performance optimisation.

One of the biggest obstacles to making evidence-based decisions on a portfolio has traditionally been the cost of quality data collection. Asset managers who want to harness data science and use lifecycle modelling have been held back by a lack of reliable data sets.

Using bad data is definitely not the answer. It leads to inaccurate forecasting and prevents proper planning and budgeting. You can’t make good business decisions using old, incomplete, or poor-quality data. There may be cost implications or a general lack of understanding for the return on investment of extensive data collection.

So, what can be done instead? 

We recently spoke at the AMPEAK conference about this issue. If intensive data collection isn’t possible, how can organisations break the negative data cycle and achieve effective lifecycle modelling?

The good news is that there are much cheaper and more efficient ways to get accurate, relevant data about assets: we call this AssetFuture Simulate.

The idea is simple: if you have a repeatable space, such as a classroom, corridor or bathroom, you can estimate the assets you’ve got by replicating that data.

It doesn’t work for unique and complex spaces such as a plant room. If it’s the only room of its kind in a building, its data can’t simply be replicated and modelled. But for the majority of spaces, it’s very effective - just consider how many essentially identical rooms there are in most premises.

At AssetFuture, we’ve been building degradation models for the last 20 years. These are combined with comprehensive asset registers and data science. Using these, we can replicate entire portfolios in just hours - without any in-field data collection - and create representative lifecycle models. It’s a reliable and powerful "short cut" to data maturity.

Creating this base level of information for AssetFuture Simulate, we conducted regression analysis on 58,000 spaces across education, health and aged care facilities. The aim was to understand the proportionality of materials and equipment in bedrooms, classrooms and shared areas.

We grouped spaces into different functional types, as well as by size. This let us translate the replacement cost ratios of materials and equipment into different ratings. We then scaled these models to a specific portfolio, based on existing floor plans. The result is holistic financial forecasting, without having to gather huge amounts of new data.

Through this project we’ve increased our understanding of the spaces in different sectors. For example, the education sector tends to have more uniform spaces, with classroom sizes and equipment being standardised across many schools. Office spaces and corridors are also relatively standard. Specialist learning spaces tend to be much more variable and do need extra data collected to increase accuracy.

In conclusion, this approach is a great way to fast-track data maturity. It "fills the gaps" in an asset register in an accurate, useful way. Asset managers can use this modelling to make much more informed decisions and become more lifecycle enabled.

AssetFuture Simulate is currently in beta as we are testing and refining the derivation process and defining the minimum viable data needed to generate a lifecycle model. Our aim is to refine the approach and incorporate machine learning to enable an accessible fast-track to data maturity.

If you would like to join the beta program for AssetFuture Simulate, please contact us.