In the case of cooling or warming devices, their usage is strongly correlated with external temperature. If this variable is known one could propose a stochastic model dependent of the temperature. For example, the hotter it is and the more likely an appliance such as a fan or an AC is to be turned on.
The external timeseries should be provided at the usecase level and the appliances could get access to it. The Appliance class would need to add a few attributes to enable dependence of an external timeseries and a way to provide the temperature dependent probability distribution.
In the case of cooling or warming devices, their usage is strongly correlated with external temperature. If this variable is known one could propose a stochastic model dependent of the temperature. For example, the hotter it is and the more likely an appliance such as a fan or an AC is to be turned on.
The external timeseries should be provided at the usecase level and the appliances could get access to it. The
Applianceclass would need to add a few attributes to enable dependence of an external timeseries and a way to provide the temperature dependent probability distribution.