From maintenance to asset management

For our machinery we are moving towards a system of asset management. We manage and guard the whole life cycle of our assets. That way, we evolve towards reliable installations at the best possible cost.

Maintenance strategy

Brecht Kestemont, department head Machinery: "Our maintenance strategy has four levels. The reactive level includes traditional repair and maintenance. Time-related maintenance is located on the preventive level. We carry out predictive maintenance based on data and sensor measurements. The final level consists of Reliability Centered Maintenance (RCM). With this method, we optimise the maintenance and augment the reliability of our machinery."

For every machine, we propose a strategy and an action plan in consultation with our internal partners - distribution centres and production sites where we cut our cheese or roast our coffee, for example. The focus is not the same for every installation. Sometimes the operational cost is more important, sometimes the availability. Of course, we are flexible enough to respond to sudden changes. That way, we keep the chain alive if higher productivity or tighter cost control is needed.

From subjective to objective

In the past, we often worked by feeling on the operation of a machine. That does not always correspond with the data. Brecht: "That is why we are removing that subjectivity from our criticality ranking - what is critical and what is less so? For that, we are not only counting on the data generated by the machine: we combine those operational hours, temperature and vibration sensors with work reports and data about interventions of our technicians. This gives us a more complete picture of how the installation operates and where we can expect problems."

"More and more, we are making use of ‘data-driven maintenance’. That data is progressively determining our maintenance strategies. Today, we are already using temperature, vibration and auditive measurements to determine the potential failure interval. We want to make our machines even smarter with extra sensors and components that generate data."

Ready for the future

In the next phase, we want to evolve further towards IoT (Internet of Things) and artificial intelligence. By doing so, we want to use correlations in the data to analyse the causes of failure. "In that way, we are moving towards machine learning. We are taking everything to a higher level to automatically sound the alarm and plan things in an action-oriented way. That way, we can send people even before machines and production chains let us down: predictive maintenance at its best", Brecht concludes.

Interested in working at Colruyt Group Technics & Engineering?