My background and education is in robotics and industrial automation.
I spent 20+ years programming, troubleshooting and improving robots and machines. Now I am more in the IT space between where stuff is made and how to improve the efficiency of the entire operation from supply chain to production to inventory and shipping logistics. My primary talent is in data querying, aggregation, analysis and visualization.
In order to solve a problem, you have to start from a assumption that all of the data that the current problematic model is based on is suspect. Then you go about validating and revising until the model's predictions reflect the observed past. Then you move forward with the new model always knowing that it's still not likely perfect and acknowledging that future adjustments are likely.