The first time we were asked to exchange with someone about Product Lifecycle Management (PLM), we needed a whiteboard within 5 seconds of discussion to explain why we should use part masters, part versions, documents. How is this connected to a product definition? How do you relate it to a manufacturing process, to a DFMEA? We ended up with lots of graphs, which once they would be validated, would start defining the digital thread for a company.
PLM backbone is about maintaining a graph
Everything is connected in a graph and the information has value for the product whenever it happens. Whether it’s a requirement or a field incident it has a lot of value to a common product. These should be easily linked.
Importance of Semantic
Storing data like you use it is the real deal. On a graph you want to be able to specify node type names but also edges names to have a graph semantic that means something for anyone looking at the data.
Further analysis with graph algorithms
In addition to creating a PLM graph for the digital thread, an interesting use-case for Graph is to analyse production incidents and look for similar causes of defects. During graphconnect 2018, Boston scientific with graphaware presented some studies they made to find product defects coming from a similar production machine. When you see where to look, than any database solution could help you. When you don’t know where to look then graphs are opening a much wider spectrum of analysis.
What about Ganister?
We do store all our data within a graph database. Which means all the data you store and manipulate has the capability to be queried with a very advanced query model with graph algorithm. It’s also much easier to understand the data. The semantic graph representing your data gives you a good understanding of your information value. API First and graph database = the right way to digitize the product lifecycle management.