Data Science in Manufacturing
Artificial Intelligence increases adherence to schedules
The manufacturing industry has the advantage of a lot of data available in many systems. The disadvantage, however, is that the collected data is usually far from practical. The consequence is very inaccurate planning, which makes it difficult to give customers exact delivery dates.
Machine learning improves planning values in machine and systems engineering
Machine and systems engineers are familiar with this problem. The very complex production planning of purchased parts and in-house production with many variants is a difficult challenge even without inaccurate planning values. The use of machine learning helps
achieve strong improvements, for example to determine replacement times more reliably and precisely. This is because the data often does not correspond to reality; production planning, for example, is then calculated to be 10 instead of 20 days. The result is very imprecise delivery dates and frequent delays. These errors can be eliminated by data cleansing using machine learning and the quality of master data therefore improved in the long term. Historical SAP data on suppliers, materials or orders, among other things, help to improve replenishment times by up to 42 percent.
The precise planning values can then be integrated into a mobile, dynamic shop floor management system where productivity key figures can be viewed directly in production at each machine via a dashboard. In this way, production can also be made digital and paperless.
Data Quality Disasters
Data quality is an innocuous term. Upon first encounter, the association is usually big tables filled with numbers, some of which erroneous, math, and complex statistics. The consequences, however, can be very real. In my previous article “Data Cleaning: Pitfalls and solutions” I shed some light on some of the shapes data quality issues can take. I also talked about a few approaches towards improving data quality and shared some insight on the business impact of inadequate data quality. Today, [...]
