Strategic Asset Management

Optimise maintenance and rehabilitation interventions in irrigation districts.

Machine learning to identify maintenance service levels and minimise costs

Like all companies engaged in water processing and delivery, irrigation districts have a large asset base. With the advent of irrigation modernisation programmes those assets are increasing as are the maintenance costs. At the same time revenue linked to water consumption is declining.

The old way of just replacing assets at end of life is no longer economically viable and a new approach based optimising costs through the asset life cycle, is now imperative.

Managing assets strategically means ensuring that asset maintenance interventions occur at the right time to return an asset to, or keep it at, the right condition to provide the right level of service. No longer do you just react to condition assessments or field reports of unserviceability.

Costs are optimised because you only maintain or refurbish to the level that is required, not necessarily back to “as new”. Costs are optimised because you intervene based on a knowledge of the life cycle cost curve of each asset type. For example intervening when condition is only slightly deteriorated is usually less costly than when condition is more deteriorated.  

The starting point for strategic asset management is to define the levels of service and risks for the company. Then you collect asset information in order to define classes of assets with common lifecycles or how service deteriorates as a percentage of remaining life. Next you identify all the types of intervention from acquisition, maintenance, performance monitoring and disposal, their costs and their effect on improvement in service level. Then you make an assessment of condition of all assets and their replacement value, and finally an assessment of the costs of failure of an asset, the risks.

With all this information you can now try a few scenarios for when to intervene and see the impact on service level, asset condition & risk over time and chose the scenario which minimises cost.

There are a number of excellent tools to automate the last step, optimising spend and providing options for decisions. But without the hard preliminary work of data collection and categorisation, they are of little use. This is where ADASA, with our experience comes in.

For one client we were able to move from a situation of annual cost blow-outs in expenditure with no end in sight, to a state where the required capital expenditure over 30 and 50 years to achieve the outcomes of risk and service could be confidently known. Then we were able to identify a range of funding alternatives and pricing mechanisms to help them move forward.

Using a different approach, another client had us use machine learning to look at all past asset maintenance activities to identify which ones in reality impacted service levels. This client wanted a more agile and targeted approach to asset base expenditure and service level differentiation across their assets, in order to reduce their asset profile to one that would be both affordable to their customers and service-appropriate as agreed by their customer base. A number of activities were surprisingly found to be both expensive and unnecessary and led to savings.


  • Reducing asset costs and risks and improving customer service by identifying asset lifecycles, conditions and valuations.
  • Optimising maintenance and rehabilitation interventions to achieve corporate objectives for level of service and risk.

Related Technologies

See our work


Contact our experts