An Adaptive Priority Approach for Effective

Problem Resolution in ITSM

 Establishing synergy between network infrastructure layer & service management layer
is critical to improve IT operational efficiency and reduce costs

SERVICE MANAGEMENT OPERATIONS MANAGEMENT

Service-Oriented Infrastructure-Oriented

Discovered CIs
Incident

Servers Applications Storage

Service-
Events & Alerts
Aware
CMDB
Problem Virtual Storage

Change virtualization Cloud Network

Service Catalog

Networks events & alerts are continuously monitored using network monitoring tools, and only the escalated alarms are logged in the service
management tool. The change management module helps plan future changes and execute any corrective actions in the network.

Copyright 2017 Prodapt Solutions. All Rights Reserved. 2

Confidential & Restricted
 Typical challenges faced by CSPs with traditional ITSM system

Growing customer base & Difficulty in identifying the Identifying the number of affected
changing needs signals out of noise customers & impact intensity

With the exponential growth in Isolating a nuisance or a noisy Identifying the accurate number of
a number of incidents, the key event can be a challenge. end users/customers affected due to
challenges faced by the Increased noisy events often network incidents is a real challenge.
operations team are identifying cause delays & visibility issues Monitoring & operations tools are
the high-priority issues, when for critical events, leading even unable to provide this essential
to address them, and whom to to customer churn. information.
assign them.

There is always a need Only reactive approach High MTTR leads to high
for an expert system - discovering problems OPEX
when users complain
Support teams have a real Currently a majority of the time is
challenge in raising multiple Inability of the current system spent in identifying & isolating a
tickets and following up on to come up with predictive & problem, rather than fixing it.
emails. There is a constant need proactive measures to prevent Hence higher MTTR (Mean Time To
for an expert system that can the incident from happening Resolve) leads to higher
automatically categorize, operational cost.
correlate & assign appropriate
severity to an incident.

A solution to address these challenges is adaptive incident prioritization model

Copyright 2017 Prodapt Solutions. All Rights Reserved. 3

Confidential & Restricted
 Need for adaptive priority model in ITSM incident management

IT & network systems have seen major evolutions during the last few decades. Strategies that helps CSPs in Adaptive Priority Model implementation
With tremendous changes in technology & services growing in leaps and bounds,
monitoring applications are unable to fulfill the dynamic demands of response
time and stringent SLAs. The number of incidents and alarms generated in the IT & Mapping business services to infrastructure
network landscape of a CSP is constantly on the rise. It is important to prioritize components
the incoming incidents with the right priority based on various attributes. However Map each business processes directly to
most ITSM systems are still prioritizing the incidents only based on urgency & infrastructure to identify impact on any service
impact. disruption. This factor improves relation between
service item and your affected incidents.

Ingest events from multiple sources across IT

landscape - Cloud enabled centralized
management system
Performing cross-domain analysis helps in big picture
view to a problem with deep dive analysis when
compared to localized analysis. Overall it improves
fault isolation and adaptive prioritization.

Adaptive incident prioritization model Real-time dashboard view for dynamic

business services
Move from traditional impact vs. urgency model to considering diverse Know the status of critical business services
attributes like number of users impacted, areas affected, services affected etc., through a single dashboard view of near real-time
to determine the right priority to an incident. view of prioritised incident.

This insights main focus is prioritizing incidents with adaptive model which helps in faster & effective resolution of a problem

Copyright 2017 Prodapt Solutions. All Rights Reserved. 4

Confidential & Restricted
 High-level architecture with adaptive priority model & topology analyzer plugins

Below solution diagram shows how incidents are analysed by the Adaptive priority Model (a plug-in integrated with existing ITSM) leveraging its
inbuilt Topology Analyser to assign right priority

ITSM-based Adaptive priority Machine learning (ML)

Incident ticketing tool incident model
algorithms to analyze various attributes
Logged With plug-in using ML
the Right
Priority
Customer Service
CMDB Topology Analyzer Class Types
(CI relations) plug-in
No. of
Customers
Affected
Primary, trigger and symptomatic
alarms are identified & correlated in
NMS NMS layer Severity Model Attributes (sample):
Business Area: Residential | Service: Voice| Type: Critical Service
Alarms enriched with info like latitude, Total Number of Users affected: 2000 | Major Incident: Regional
longitude, domain, parent entity and Impact: Significant | Urgency: High
EMS EMS EMS
affected fields in EMS layer. Priority: 1 (Custom priority assigned by adaptive system)

Topology Analyzer Classification algorithms Adaptive Priority Incident Model

Key - Open API enables easy integration with ITSM systems - No manual selection - Multiple parameters like business area, service affected, category of
Accelerators - Intuitive GUI, drill down feature to view affected users - Algorithm-based business rules/ML service, number of customers impacted & level of service degradation
- Navigation capabilities upto tail node - Improved classification process are considered for identifying the effective priority for an incident.

Copyright 2017 Prodapt Solutions. All Rights Reserved. 5

Confidential & Restricted
 Recommendations - Prioritization of incidents based on advanced data set ..Part(1/2)

Problem Analysis IMPACT

As the number of incidents grows, the support team finds it challenging to handle them. The existing

URGENCY
incident priority calculation model using traditional method is shown in the below matrix which is based on
impact vs. urgency.

Recommended Solution

Traditional ITSM systems depend only on impact & urgency factors to determine the incident priority while Adaptive Incident Priority Model Plugin equips
ITSM systems with additional capability of looking into multiple influencing factors, which helps in determining the right priority. Without a framework for
establishing correct priority, it is very difficult to meet service level agreements & trigger incident escalations appropriately.

Enriching the incident with intelligent data helps in effective problem identification & resolution.

Adaptive Priority Incident Model solution is based on new business logic/algorithm with various input parameters which include:

Business area (residential, B2B, IT) | Service affected (local, urban, regional) | Service category (DTV, CRM, billing)
| Impacted customers (0~500,501~5000,5001~50,000)

Open API availability on CMDB or network topology database makes it easier for adaptive plugin integration.
Copyright 2017 Prodapt Solutions. All Rights Reserved. 6
Confidential & Restricted
 Recommendations - Prioritization of incidents based on number of end-users ..Part(2/2)

Problem Analysis NODE

RELATIONSHIP
A typical network is a complex system which comprises multiple telecom nodes from
different vendors across multiple domains. Accurately identifying the number of affected
end-users because of an incident is a real challenge. NODE RELATIONSHIP

For e.g., smooth operation of a wireless network depends on the underlying fiber RELATIONSHIP
network.
NODE

Recommended Solution

Topology Analyzer - an intuitive GUI, which is used to traverse the entire network of the CSP.

Based on the relationship between two nodes in a telecom network, we can identify its type of relationship i.e. either its a parent-child or uncle-nephew.
If a port does not exist without a card, it is an example for parent-child relationship.

Based on selected incident from the network, the faulty element can be identified through the Topology Analyzer which in turn helps in identifying the
number of end-users affected or the portion of customer base impacted on the specific node through its relationship data. Adaptive priority model
helps in assigning the right priority to the incidents based on this info.

Prerequisite for implementing relationship based prioritization model is the availability of relationship data of entire network

Copyright 2017 Prodapt Solutions. All Rights Reserved. 7

Confidential & Restricted
 Sample data set for adaptive priority model & Topology Analyzer view

Sample data set used for identifying

adaptive priority

Business Area Service Impact Impacted Customers Service Category Key Service Impact Urgency Priority
Residential Local 0-500 DTV Tier 2 4-Minor/Localized Low Low
IT Urban 501-5000 CRM Tier 2 2-Significant/Large Medium Medium
IT Regional 5001-50000 Billing Tier 1 1-Extensive/Widespread High High
Nationwide or Multiple
B2B Countries 50001-999999999 Customer-facing Mail B2B Tier 1 1-Extensive/Widespread Critical Critical

Topology analyzer shows number of end users affected

because of an incident

Element ID
Type of Customer DTV Internet Voice B2B
1611E- 08-008 No. of affected customers 15 28 26 3

Copyright 2017 Prodapt Solutions. All Rights Reserved. 8

Confidential & Restricted
 Features that adaptive incident priority model brings into traditional ITSM system

Handling complex associations | Multi-technology,

multi-vendor
01 Captures & maintains all the complex associations in an
ecosystem with its relationship across different domains
E.g., relationship between DWDM equipment and
Edge router (IP based)

02 Dynamic impact assessment

Near-real-time impact assessment through in-memory
relationship data using ML-based algorithm
Resolution
research

03 In radar monitoring
Continuous monitoring of affected configuration items(CIs) in
service impacted areas

04 Proactive notifications
Incident priority-based notifications to all relevant stake
holders & the management team depending on the escalation
matrix

Copyright 2017 Prodapt Solutions. All Rights Reserved. 9

Confidential & Restricted
 Key Takeaways

Up to 15% improvement in SLA adherence because of assigning right priority to an incident

Up to 20% improved customer satisfaction due to right prioritization & quick resolution time
Up to 10% reduction in P1 & P2 average ticket logging volume
Up to 50% reduction in average time spent in analyzing & resolving the tickets

Copyright 2017 Prodapt Solutions. All Rights Reserved. 10

Confidential & Restricted
 Credits

Mogan AB, Manager Technical Pre-Sales & Insights

Kayalvizhi S, Associate Director ITSM Delivery
Omprakash C, Technical Architect ITSM Delivery
Manju P, Senior Technical Lead ITSM Delivery

Copyright 2017 Prodapt Solutions. All Rights Reserved. 11

Confidential & Restricted
 Amsterdam

London

New York
Tualatin THANK YOU!
Dallas

Bengaluru
Chennai

Johannesburg

USA UK THE NETHERLANDS SOUTH AFRICA INDIA

Prodapt North America Prodapt (UK) Limited Prodapt Solutions Europe Prodapt SA (Pty) Ltd. Prodapt Solutions Pvt. Ltd.
Tualatin: 7565 SW Mohawk St., Reading: Davidson House, Amsterdam: Zekeringstraat 17A, 1014 BM Johannesburg: No. 3, Chennai:
Ph: +1 503 636 3737 The Forbury, Ph: +31 (0) 20 4895711 3rd Avenue, Rivonia 1. Prince Infocity II, OMR
Reading RG1 3EU Ph: +27 (0) 11 259 4000 Ph: +91 44 4903 3000
Dallas: 222 W. Las Colinas Blvd., Irving Ph: +44 (0) 11 8900 1068
Prodapt Consulting BV 2. Chennai One SEZ, Thoraipakkam
Ph: +1 972 201 9009
Rijswijk: De Bruyn Kopsstraat 14 Ph: +91 44 4230 2300
Ph: +31 (0) 70 4140722
New York: 1 Bridge Street, Irvington
Ph: +1 646 403 8158 Bangalore: CareerNet Campus
No. 53, Devarabisana Halli,
Outer Ring Road