Which machine learning technique can be used for network downtime prediction in telecom




I would like to know how is the network down phenomenon predicted in telecom sector.?Is some form of mapping is used(like ggmap) to pin point the places where it is going to occur,or is something else done for this.What package in R can be used for this.Also I would like to have some data to work on such a problem,so if any leads can be provided on from where I can get such data to work upon it would be very helpful.