Please provide some direction for the problem I am trying to solve.
IoT sensors at villas generate Fire and Maintenance alamrs. I have alarm history data - Customer no., timestamp of alarm, problem category etc. Along with this I have Customer number, customer details, location, villa details, number and type of smoke, heat, repeaters sensors.
It is easy to predict probability of customers having a particular type of maintenance alarm(using classifier - yes/no). It is easy to prepare data Datewise and do forecast of the no. of alarms using regression(Date wise forecast).
The problem I am trying to solve is, a step above this. I need to do “Datewise” (for next week in future) forecast for the Customer numbers who are expected to have that particular problem category of alarm.
Ultimately we want to provide a set of customer numbers for each day in coming future week, which may have this alarm. And we want to do planned maintenance so that we can fix the problem before it occurs.
Any pointers will be a great help.