Abstract: In some implementations, a device may provide, as input to a model, input data indicating current value(s) corresponding to key performance indicator(s) (KPI) associated with current calls to a contact center, enterprise event(s), and stability event(s) associated with the contact center. The device may receive, as an output from the machine learning model, predicted value(s) corresponding to the KPI(s). The device may determine predicted rate(s) of change of the KPI(s) based on a comparison of the current and predicted value(s). The device may assign severity level(s) corresponding to the KPI(s) based on the predicted rate(s) of change. The device may determine a service level impact based on the predicted value(s). The device may determine a performance degradation score based on the severity level(s) and the service level impact. The device may determine and automatically apply a strategy based on the performance degradation score.