First, let's look at the role of data in optimising industrial efficiency and productivity. Many industrial customers are literally sitting on mountains of data, which is not always utilised. This is not always a bad thing, as historical data in particular can have little relevance to today's challenges. The market situation, legislation, regulations, new products and customer behaviour are constantly changing. It is therefore important to first find out what is relevant and what is not. We like to call this task ‘finding the signal in the noise’. In order to find the signal, we have to make the data accessible, which is often found in many separate sources (data silos).
So what does this look like in practice? In the following, we use a real-life example to show how a customer from the energy sector was able to improve processes with the help of analyses:
The customer operates wind turbines. The challenge was that faults were only rectified reactively. This cost money twice over: on the one hand, the cost of the repair itself and, on the other, the opportunity cost of the lost electricity production for the duration of the repair.
We were able to help the customer by using available fault data to develop patterns, so-called lead measures, which predict that a fault will occur. Thanks to our prediction, the customer can now predict at least 80% of incidents and thus plan maintenance - e.g. by postponing it to windless days. This saves him double the costs.
Is real-time monitoring always necessary to gain predictive insights?
The answer is that real-time monitoring is not always necessary, for example when planning demand from B2C customers. Real-time monitoring is particularly important in companies with continuous operations such as power generation and metallurgy.
What are the challenges and solutions for data transfer, especially for traditional industries?
Your challenge: You have data, but don't know what to do with it.
The solution: Start small with a workshop and a clearly defined use case.
Your Challenge: no time
Solution: Start with a workshop for a specific use case as shown above, as this allows you to build a proof of concept with little time and determine whether a larger project is worthwhile.
Your challenge: The data is scattered.
Solution: Data Engineering is the answer - a small virtual Data Mart can draw on only the necessary data to be analysed without burdening large systems such as SAP or DWH.
We at Datasense Consulting deal with this topic. We are happy to help you realise the full potential of your data.
Get in touch with us - we will be happy to advise you!
Comments