In this chapter, we reinforced our decision science learnings by solving an altogether new use case for the solar energy industry. We started on the same roots of problem solving by defining the problem and designing the approach and blueprint for the problem. We studied that the problem statement in our use case is much more specific and narrowed-down compared to the previous use case. We solved the problem of uncertainty in power outages from the solar tech. After having a clear definition of the problem statement, we explored the sensor data from solar panels and infrastructure to find patterns and associated signals. After gathering the ground context of the data and domain (through research and an SME), we engineered features to solve our problem better.
We then leveraged these features and the machine learning algorithms that you learned in the previous chapter to build predictive models that could predict the chances of power outage from the solar panel infrastructure a day...