Abstract: In some aspects, the techniques described herein relate to a method including: identifying a first base station (BS) that is out of service at a first time; identifying a first set of user equipment (UE) connected to the first BS in a pre-outage time window occurring before the first time; identifying a second set of UEs, the second set of UEs including UEs connected to other BSs in a wireless network; computing a set of seen UEs based on the second set of UEs and the first set of UEs; computing a set of unseen UEs based on the set of seen UEs and the first set of UEs; predicting a set of UEs that have left a service region that includes the first BS and the other BSs; computing a set of UEs without service based on the first set of UEs, the set of seen UEs, and the set of UEs that have left the service region; and generating a visualization of the set of UEs without service.

Abstract: A device may determine a number of user equipments (UEs) connected to an inactive site during a first period of time and determine that the inactive site experienced a power outage during a second period of time. The device may determine that the UEs connected to a set of active sites during the second period of time. The device may determine one or more key performance indicator (KPI) changes to one or more KPIs, during the second period of time, for each of the set of active sites. The device may determine a plurality of weights associated with the inactive site experiencing the power outage; determine a set of impacts, of the inactive site, on the one or more KPI changes for the set of active sites based on the plurality of weights; and cause power to be restored to the inactive site based on the set of impacts.