Predictive asset management

Asset management excellence by data driven insights

Asset-intensive industries such as energy & resources, manufacturing, ... continuously strive to optimise the performance of their physical assets. As a result, many asset managers are facing challenges such as the optimisation of investments and operations, the allocation of scarce capital in the most optimal way amongst the full range of assets, the impact of investment decisions made on the strategic objectives of their organisation, ... To address these challenges, Deloitte developed a predictive asset management approach of which the "intelligent Asset Control Center“ is the ultimate solution. This innovative solution enables asset intensive companies to use a data-driven approach to make better decisions, gather deeper insights, enhance strategic asset investments, and drive a more effective operations planning process.

Find below the description of the predictive asset management 5 steps approach!

1. It all starts with data

Deloitte offers a 360° view on assets, taking into account data about technical characteristics (such as type, vendor, material or construction date), operational characteristics (such as maintenance history or events), measurement data (such as condition monitoring from telemetrics) and environmental data (such as temperature, humidity, soil type and wheather data) available from their scada, their asset management systems, etc.

2. Root cause analysis (FMEA) and failure pattern recognition

Once all data data is integrated and the quality is sufficient (typically tested by means of a feasibility assessment), a root cause analysis (FMEA) and failure pattern recognition is being performed to understand past behavior, this being an analysis of historical data to create insights into the failure behavior of assets.

3. Predictive and statistical modelling

After looking at the past, knowledge is being created by predicting the future in terms of asset health. This means that predictive and statistical models based on survival analysis are being used to calculate likelihood, represented by amongst others the mean remaining lifetime within a particular confidence interval and this for each individual asset.

4. The asset health index

The knowledge obtained is turned into decisions by establishing an asset health index (AHI). The predicted lifetime serves as input for the asset health index representing the probability of failure. However: this is not the only important parameter to take into account when determining asset health, also the impact is as important as the predicted probability. Based on this AHI, alarm generations can be generated and maintenance plans or capital allocation can be determined.

5. Integrating decisions into actions and work orders

The final phase relates to integrating the decisions taken into actions and work orders and thus integrating the iACC approach within the real business processes of a company. This directly impacts the current operating model and therefore change management is a critical success factor. That is why iACC is not only a technology solution or just a predictive model, but a business transformation programme.