The current situation and high-level findings
Pilots & strategy
The pilots were stand-alone initiatives with their own limited scope. They were a great way to gain more knowledge and understanding of data driven maintenance, the tools and systems and opportunities available. However, many initiatives got stuck before reaching an improvement and most forgot to share learnings with the rest of the organization. The strategy or objectives of the pilots were not always clear, nor was the alignment with the (long-term) business strategy. In addition, every pilot team tried to find their own way through the ICT infrastructure.
Asset failure history
The available failure data did not provide enough insights to be able to pinpoint suitable asset types for data driven maintenance. This is also a challenge for the future as predictive models need to be trained based on historical failure data. Vitens is already working on solving this as they set up their asset register in a new version of their EAM system with a better setup. In a few years they will be able to have better data insights.
Asset base, cost and maintenance
Vitens aims to ensure a very high availability, which is mainly achieved by having a lot of redundancy in their asset base. As all of these assets need to be maintained, their maintenance costs are relatively high. Therefore, looking from the perspective of the asset base, optimization of maintenance is most suitable for the goal of cost reduction. In the case of Vitens, reducing cost is not achieved by reducing maintenance scope, but by optimizing the moment of maintenance execution through setting up data triggers.
To give an example: reservoirs are in the top 3 of asset types with the highest costs. For these assets periodical cleaning is needed to ensure water safety and hygiene. This takes a reservoir out of commission for 24 to 48 hours which is, of course, costly. Predicting when the reservoirs need maintenance can improve planning and preparation efforts which can be more efficient and cost-effective. Hence, reducing the asset scope is not an option, but the optimization of maintenance execution is.