Despite all the discussion about digitization and Industry 4.0. Production is rich in data but poor in information. It is not uncommon for a perceived minimum of productivity data to be recorded via imprecise handwriting. The potential of modern “manufacturing intelligence” is not being used. But it's not that difficult at all.
Even old control systems record a basic amount of data that provides information about the status and potential of productivity. Modern systems naturally offer many more options, but more is not automatically better. The news is that data is there and the problem of the lack of connectivity between the systems has been solved. It is now possible for almost every system to obtain data automatically and at very manageable costs. The time for handwriting is over.
In IT terms, what a connected system then generates is time series data. In this case, the process parameters for the respective time unit are stored in small time increments in a special database. This could be, for example, quantity, percentage of feed or scrap.
With just a few measurement parameters, the OEE can then be displayed in real time. Once this technical and informational basis has been laid, the infrastructure can be supplemented. To enrich the data stream with further information, you would “annotate” it. Which production order is part of this time increment? , what was the reason for the standstill? This informational addition can be provided by other IT systems, such as machine sensors or even by humans.
The resulting “enriched” time series is then in turn the basis for analysis using algorithms. This type of data storage enables rapid access and analyses in real time. Whether these are statistical algorithms or artificial intelligence depends on the question at hand. Both are possible. There is also no limit whether questions from the past should be answered — such as finding clusters of certain faults in the data stream — or whether a statement should be made for the future — for example when the next malfunction is to be expected. The technology allows for all application scenarios.
For some questions, an algorithm must first be trained. In these cases, the question clearly belongs to the field of artificial intelligence. It is also understandable that a time series must first be built up to use the algorithms. This data collection period depends on the processed data and can take from a few weeks to many months. Only when there is a sufficient history can an AI algorithm provide reliable information.
In reports and on Andon boards, the information can be displayed immediately after installation and configuration. Once the described path has been completed, the step towards push notifications in the event of anomalies is only a short one. For example, faults can then be communicated to those responsible while they occur. The reaction speed increases, so does productivity.
This is modern “manufacturing intelligence”: Analyzing plant productivity based on real-time data, making information available to specific employees and increasing OEE through targeted measures. Please feel free to contact us.