Contextual data holds much more promise than data lakes for 2 reasons: Firstly, most high-ROI activities in discrete manufacturing are based on knowing what happened to the part. The power of knowing what order the part is related to, what material genealogy looked like, what the machine did during production and what operator handled it, when, and how long makes as difference for the part. Building on this knowledge with new software tools doesn’t only yield immediate improvements in process efficiency and reliability, but also enables additional, value-added tools. These could include benchmarking, supplier management, sustainability calculation, quoting, toolpath planning and much more.
Secondly, these value-added applications and process improvements benefit from an underlying trend: The increasingly algorithmic nature of engineering software. Previously, moving parts from designing through manufacturability and into production required the use of dozens of different pieces of software and interfaces which were complex and expensive. Instead, algorithms for complex algorithms such as mesh healing, orientation, support, nesting, quoting and even simulation and designs are beginning to be available as API’s, hidden from the user view, easy to exchange, and in-line to any process. The industry must adopt these trends if it is to meet the demand for more agility emanating from customers and markets. The additive manufacturing industry has already started to experiment with the resulting opportunities. With their experience, and a focus on contextual data, lot size one at the cost of lot size 1bn seems no longer out of reach.