Volvo Cars is moving towards an increasingly data-driven existence
Working with data-driven decisions is one of 4 central principles of Smart Maintenance. With today's advanced production equipment and technology specifically for data collection, the possibilities are almost endless. One challenge that many companies encounter, however, is that they collect large amounts of data that you then do not really know what to do with. Or the stored data is difficult to extract and compile from (often many different) systems.
We began to map our workflows around data analysis and realized that we spent an unreasonable amount of time just compiling the data needed in our analyzes.
– Krister Mattsson, Smart Maintenance Engineer in the body factory at Volvo Cars, Torslanda.
One of the challenges at Volvo Cars was that the data was stored in many different systems, which resulted in the work process for collecting and analyzing data being very time-consuming.
- Our specialist technicians spent 3.5 hours just collecting data before they could start analyzing what has actually happened, which is not at all using our resources correctly.
Krister tells specifically about an example to follow up quality deviations. There, they have previously spent a lot of time verifying daily that they are within the tolerances. Now they instead get an alarm when it goes outside the tolerances, and a work order is automatically generated.
- Technically, it is nothing new in maintenance, but we have worked hard to make it easier to extract data to make the analysis and present it in a way that is useful for our technicians, says Krister.
They have a continuous dialogue with the maintenance technicians and specialists to ensure that the tools that are developed are useful and contribute to a working method that lasts in the long run.
Much of the data collected today is stored in central databases and an effort has been made to get the data more accurately timed. For example. if alarms are logged at a second level, there is a risk that several alarms will receive the same timestamp, which makes it difficult to know which alarm came first and thus complicates the work with root cause analysis. If you instead log the alarms at micro- or nanosecond level, it gives greater opportunities to find the root cause based on the order in which the alarms actually occurred.
At Volvo Cars, many initiatives are underway to go from calendar-based FU (preventive maintenance) and patrols to maintenance based on operating hours / cycles and alarms triggered by simple algorithms. Krister takes leakage of compressed air as an example: in the past, technicians have toured the entire factory to check for leakage of compressed air. Now they have installed technology that helps regulate the pressure and detect leakage. The pressure is regulated down (the air is temporarily shut off) during production stoppages, such as breaks, instead of continuing to maintain the pressure required during production. If there is a leak, the pressure will go down quickly. Should the pressure reach a critical level, the air is switched on again to prevent damage to components and production equipment. In this way, much less compressed air is used during production stoppages compared to continuously maintaining production pressure despite leakage, and Volvo saves money.
- Now that we have started measuring, we can see that we have much more leakage than we previously thought. So we save more money than we expected when we did the evaluation of how much savings we could make.
In addition to saving money on reducing the use of compressed air, the technicians work much more efficiently when they do not have to go out and look for leaks. An automatic work order is now sent with information about how much it is leaking. Soon they also want to be able to send text messages with work orders to individuals for even faster action.
So far, they have focused on working with the data that they 100% know the technicians need in the work towards reducing the number of urgent cases - the data they use but have had to spend a lot of time producing. In the future, they also want to work with machine learning, and projects with external parties are being started up. However, a central initiative is needed to set the standard for working methods.
- We want a global standard with common working methods at all our factories. In Torslanda, we have already tried new technical solutions to support our technicians, but to move forward in the development, the initiative needs to be raised centrally at Volvo. What is developed must be able to be implemented globally.
Finally, Krister wants to emphasize the importance of defining and continuously working with goals.
In order to succeed with data-driven working methods, a clear plan for the organizational development is needed that supports the company's overall goals regarding Smart Maintenance.
– Krister Mattsson, Smart Maintenance Engineer in the body factory at Volvo Cars, Torslanda.
The article is written by Camilla Lundgren based on an interview with Krister Mattsson at Volvo Cars.
Tags
This article is tagged with these tags. Click a tag to see all the articles with this tag.