Network service incident prediction

Patent metadata

Field	Value
Publication number	US-9806955-B2
Application number	US-201514831168-A
Country	US
Kind code	B2
Filing date	Aug 20, 2015
Priority date	Aug 20, 2015
Publication date	Oct 31, 2017
Grant date	Oct 31, 2017

Abstract

A device may obtain first information related to network devices of a network. The device may obtain second information related to the network devices and/or to one or more historic network service incidents. The one or more historic network service incidents may be related to network services provided in association with the network devices. The one or more historic network service incidents may include outages and/or degradations of one or more network services. The device may perform an analysis of the first information and the second information. The device may train a predictive model based on the analysis of the first information and the second information. The predictive model may predict a probability of a future network service incident based on the first information and/or the second information. The device may cause third information, related to the network devices, to be monitored based on the predictive model.

First claim

What is claimed is:

1. A device, comprising:
a memory storing instructions; and
one or more processors to execute the instructions to:
obtain first information related to network devices of a network,
obtain second information related to at least one of:
the network devices, or
one or more historic network service incidents associated with the network devices,
the second information being different than the first information,
the one or more historic network service incidents being related to network services provided in association with the network devices, and
the one or more historic network service incidents including at least one of:
 outages, or
 degradations of one or more network services;

parse the first information and the second information;
structure, based on parsing the first information and the second information, the first information and the second information to generate structured first information and structured second information;
normalize the structured first information and the structured second information to generate normalized first information and normalized second information;
perform an analysis of the normalized first information and the normalized second information;
train a predictive model based on the analysis of the normalized first information and the normalized second information,
the predictive model predicting a probability of a future network service incident based on at least one of:
the normalized first information, or
the normalized second information; and

cause third information, related to the network devices, to be monitored based on the predictive model.

2. The device of claim 1 , where the one or more processors are further to:
determine that a first portion and a second portion, of at least one of the first information or the second information, relate to particular data,
the first portion including different information than the second portion; and

assign a value, to the first portion and the second portion, based on the first portion and the second portion relating to the particular data,
the value identifying the particular data.

3. The device of claim 1 , where the one or more processors, when causing the third information to be monitored, are further to:
receive the third information; and
process the third information to associate the third information with the normalized first information and the normalized second information.

4. The device of claim 1 , where the first information relates to a configuration of the network devices; and
where the second information relates to operation of the network devices; and
where the third information relates to the operation of the network devices,
the third information being monitored after the normalized second information is obtained.

5. The device of claim 1 , where the one or more processors, when performing the analysis, are further to:
perform a chi square automatic interaction detection analysis to train the predictive model.

6. The device of claim 1 , where the one or more processors are to:
predict a network service incident, with regard to one or more of the network services, based on the third information and the predictive model; and
provide a notification to one or more other devices based on predicting the network service incident.

7. The device of claim 1 , where the one or more processors, when training the predictive model, are to:
determine a first criticality coefficient and a second criticality coefficient,
a network service incident that is associated with a relatively higher cost being assigned the first criticality coefficient, and
a network service incident that is associated with a relatively lower cost being assigned the second criticality coefficient;

determine a first cost associated with a first network service incident and a second cost associated with a second network service incident,
the first cost being a higher cost than the second cost;

assign, to the first network service incident, the first criticality coefficient; and
assign, to the second network service incident, the second criticality coefficient.

8. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors to:
obtain first information related to network devices of a network,
obtain second information related to the network devices and to one or more historic network service incidents associated with the network devices,
the second information being different than the first information, and
the one or more historic network service incidents being related to network services provided in association with the network devices;

structure the first information and the second information to generate structured first information and structured second information;
normalize the structured first information and the structured second information to generate normalized first information and normalized second information;
perform an analysis of the normalized first information and the normalized second information;
predict a business impact, related to a future network service incident, based on performing the analysis of the normalized first information and the normalized second information;
generate a remediation plan, based on the business impact and related to the network devices, to reduce a cost related to the future network service incident; and
provide information related to at least one of:
the remediation plan, or
the business impact.

9. The non-transitory computer-readable medium of claim 8 , where the one or more instructions, that cause the one or more processors to predict the business impact, cause the one or more processors to:
determine a historic cost related to the one or more historic network service incidents;
determine a relationship between the normalized first information, the normalized second information, and the historic cost; and
predict a future cost related to the future network service incident based on the relationship between the normalized first information, the normalized second information, and the historic cost.

10. The non-transitory computer-readable medium of claim 8 , where the one or more instructions, that cause the one or more processors to predict the business impact, cause the one or more processors to:
identify a location that is associated with one or more of the network devices and one or more of the network services,
determine a cost associated with an outage at the location,
the outage at the location being associated with one or more network service incidents related to at least one of:
the one or more of the network devices, or
the one or more of the network services; and

where the one or more instructions, that cause the one or more processors to generate the remediation plan, cause the one or more processors to:
generate the remediation plan based on the cost associated with the outage,
the remediation plan identifying one or more actions to perform with regard to at least one of:
the location, or
the one or more of the network devices.

11. The non-transitory computer-readable medium of claim 10 , where the cost is a first cost; and
where the location is a first location; and
where the one or more instructions, that cause the one or more processors to determine the first cost, cause the one or more processors to:
determine a second cost associated with a second location,
the second location being different than the first location, and
the second cost being a lower cost than the first cost; and

wher

Assignees

• Accenture Global Services Ltd

Inventors

• Duncan Peter T
• Sareen Kundan Lal
• Advani Jai
• Chakraborty Amab D
• Nair Vinita V
• Venkataraman Vinoth

Classifications

• H04L41/149

  for prediction of maintenance · CPC title

• H04L41/147

  for predicting network behaviour · CPC title

• G06F11/008

  Reliability or availability analysis · CPC title

• H04L41/145 Primary

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/16

  using machine learning or artificial intelligence · CPC title