Storm response optimization

Abstract

A method of predicting equipment failures is provided. Potentially-effected equipment located in a path projection for a weather event is identified based on a current characteristic data describing equipment supporting a service. A likelihood of failure for each equipment of the identified potentially-effected equipment is calculated by executing a failure prediction model with the current characteristic data and weather event data describing characteristics of the weather event. Equipment failures of the identified potentially-effected equipment are predicted by comparing the calculated likelihood of failure for each equipment of the identified potentially-effected equipment to a predefined threshold. Information identifying the predicted equipment failures is output.

What is claimed is: 
     
         1 . A computer-readable medium having stored thereon computer-readable instructions that when executed by a computing device cause the computing device to:
 receive current characteristic data of equipment supporting a service;   receive weather event data including a path projection for a weather event;   execute a failure prediction model with the received current characteristic data and the received weather event data to define a likelihood of failure of the equipment supporting the service; and   identify predicted equipment failures of the equipment supporting the service by comparing the defined likelihood of failure of each equipment of the equipment supporting the service to a predefined threshold.   
     
     
         2 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to:
 receive data associated with second equipment supporting the service, wherein the data includes previous weather event data and characteristics of the second equipment supporting the service during the previous weather event;   determine characteristics associated with a failure of a subset of the second equipment during the previous weather event; and   define the failure prediction model based on the determined characteristics.   
     
     
         3 . The computer-readable medium of  claim 2 , wherein the characteristics of the second equipment are selected from the group including a last equipment failure data, a last equipment service date, an equipment type, an equipment age, and a last tree trimming date associated with a location of the equipment. 
     
     
         4 . The computer-readable medium of  claim 2 , wherein the previous weather event data are selected from the group including a wind speed, a rainfall rate, a snowfall rate, and a wind direction. 
     
     
         5 . The computer-readable medium of  claim 2 , wherein the characteristics of the second equipment are determined based on a degree of correlation in predicting the failure of the subset of the second equipment during the previous weather event. 
     
     
         6 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to indicate locations of the identified predicted equipment failures on a map. 
     
     
         7 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to determine an event level classification based on the identified predicted equipment failures, wherein the event level classification is a measure of a severity of expected service outages resulting from the identified predicted equipment failures. 
     
     
         8 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to determine a staging location for a service outage responder based on the identified predicted equipment failures. 
     
     
         9 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to determine an estimate of replacement equipment needed to respond to the weather event based on the identified predicted equipment failures. 
     
     
         10 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to receive outage data identifying a service outage source location. 
     
     
         11 . The computer-readable medium of  claim 10 , wherein the outage data is defined by analyzing social media data. 
     
     
         12 . The computer-readable medium of  claim 10 , wherein the computer-readable instructions further cause the computing device to indicate locations of the identified predicted equipment failures on a map and to indicate the service outage source location on the map. 
     
     
         13 . The computer-readable medium of  claim 10 , wherein the computer-readable instructions further cause the computing device to determine an event level classification based on the identified predicted equipment failures and the received outage data, wherein the event level classification is a measure of a severity of expected service outages resulting from the identified predicted equipment failures. 
     
     
         14 . The computer-readable medium of  claim 10 , wherein the computer-readable instructions further cause the computing device to determine an estimate of replacement equipment needed to respond to the weather event based on the identified predicted equipment failures and the received outage data. 
     
     
         15 . The computer-readable medium of  claim 10 , wherein the computer-readable instructions further cause the computing device to determine service restoration time estimates based on the received outage data. 
     
     
         16 . The computer-readable medium of  claim 1 , wherein the computer-readable instructions further cause the computing device to:
 receive updated weather event data;   execute the failure prediction model with the received current characteristic data and the received updated weather event data to define an updated likelihood of failure of the equipment supporting the service; and   identify updated predicted equipment failures of the equipment supporting the service based on the defined updated likelihood of failure of the equipment supporting the service.   
     
     
         17 . The computer-readable medium of  claim 1 , wherein the equipment supporting the service and the second equipment supporting the service include the same equipment. 
     
     
         18 . The computer-readable medium of  claim 1 , wherein the failure prediction model includes one or more mathematical models selected from the group consisting of a regression model, a decision tree, an ensemble model, and a neural network model. 
     
     
         19 . A system comprising:
 a processor; and   a computer-readable medium operably coupled to the processor, the computer-readable medium having computer-readable instructions stored thereon that, when executed by the processor, cause the system to   receive current characteristic data of equipment supporting a service;   receive weather event data including a path projection for a weather event;   execute a failure prediction model with the received current characteristic data and the received weather event data to define a likelihood of failure of the equipment supporting the service; and   identify predicted equipment failures of the equipment supporting the service by comparing the defined likelihood of failure of each equipment of the equipment supporting the service to a predefined threshold.   
     
     
         20 . A method of predicting equipment failures, the method comprising:
 identifying, by a first computing device, potentially-effected equipment located in a path projection for a weather event based on current characteristic data describing equipment supporting a service, wherein the equipment is land-based;   calculating, by the first computing device, a likelihood of failure for each equipment of the identified potentially-effected equipment by executing a failure prediction model with the current characteristic data and weather event data describing characteristics of the weather event; and   predicting equipment failures of the identified potentially-effected equipment by comparing the calculated likelihood of failure for each equipment of the identified potentially-effected equipment to a predefined threshold; and   outputting information identifying the predicted equipment failures.   
     
     
         21 . The method of  claim 20 , wherein outputting the information comprises indicating locations of the predicted equipment failures on a map for presentation in a display.