Data was not the missing piece.
At the 2018 Hackbay hackathon, I joined a team exploring the everyday decisions factory managers make. We had access to machine data connected to Siemens MindSphere and visited the Siemens plant in Amberg.
The problem was not a lack of data. It was turning that data into an action at the moment someone needed to make a decision.
Make experience available to the next person.
Our concept connected machine states, decisions, and their consequences. The hypothesis: documenting that relationship could help a recommendation system draw on the experience of previous decision-makers.
We designed two connected experiences. One let experts assess suggested actions to help create training data. The other presented recommendations and feedback in the context of a factory manager’s work.
A prototype, and a question worth pursuing.
We worked with Siemens mentors, tested the concept through prototypes, and developed the interface alongside the algorithm and application. The team produced a functional prototype during the hackathon.
decAIde won the Siemens challenge at the 2018 Zollhof hackathon. The project explored a shift I still find useful: moving from presenting more information to supporting a specific decision.
The original prototype established the concept; it did not establish production performance or the reliability of AI recommendations in a live factory.