de Nijs_Jan_IND cropped.png


Jan de Nijs | Tech Fellow for Enterprise Digital Production 
| Lockheed Martin

Jan de Nijs is the LM Fellow for Enterprise Digital Production and co-leader of the team within the Lockheed Martin Digital Transformation Program that has been chartered with unlocking the value of Digital Production. Before joining Lockheed Martin, he spent the first 15 years of his career in leadership roles in the capital goods industry, designing and building automation and metrology solutions for the automotive, aerospace, and the food packaging industry, working both in Europe and the United States. In 2004, he joined the Lockheed Martin Missiles and Fire Control Division in Orlando, Florida, where he worked as a leader on factory automation and production process improvement projects. In 2016, he moved to the Lockheed Martin Aeronautics Division in Fort Worth, Texas, to become part of the F-35 Production team. In 2018 he got recruited to lead the corporate effort for enterprise digital production across all Lockheed Martin divisions. Jan holds an MS degree as well as a Technical Doctorate in Mechanical Engineering from Eindhoven University of Technology in The Netherlands.


It’s all about the digital thread! Making Manufacturing Data More usable for Analytics

Surprisingly, many large corporations are struggling to create viable and profitable advanced analytics solutions using data produced in the Life Cycle chain of products. The problem is not the volume of data: with the proliferation of connected devices, we are now awash in Life Cycle data. Rather, the problem is that data created by Life Cycle Producers (sources) almost always lacks meta-data information that can be used to create a usable digital thread (missing semantics connectors). In other words, it is very hard to develop generalized solutions that can autonomously interpret Life Cycle data and create insights across the product Life Cycle.

That does not mean that the non-semantics data generated today is entirely useless, because given enough budget, impressive demonstrations can be created today. The problem is that these demonstrations are proving incredibly difficult to scale beyond the demonstration phase. Speaking about the manufacturing field, Bruce Kramer from the American National Science Foundation (NSF) once said: “at great cost and effort, we’ve been doing expensive demos in the hope that generalized principles emerge”. And generalized principles are not emerging.

This presentation explains the standards-based approach Lockheed Martin is taking to address this issue. The solution centers around a digital thread that has been defined around specific meta-data tags that allow for an automatic link back to the model-based engineering requirements, and an ecosystem of internationally accepted standards. By implementing this solution across the complete Product Life Cycle, analytics solutions can be created that have the potential to become prescriptive, or even cognitive (“the factory runs itself!”).



Why is this applicable? Manufacturing has a lot of trouble retaining skilled data scientists. The problem is that within manufacturing, data scientists are being used as data janitors (manually cleaning up data), instead of enabling them to do what they are good at: creating big  data analytics solutions.