Analysis of interventions to optimize a pumping process in food production

In this work in progress report we show the use of Causal Inference with a Double Machine Learning approach to derive recommendations to favorably intervene in a pumping process in food production. At this point the machine learning does not yield statistically significant results due to lack of sufficient data. We further look into strata we have built from the data to study treatment effects but are again limited by the amount of data available. An outline how we intend to proceed with the work is given.