Advanced Analytics for Intelligent Choices

aroocarmineΤεχνίτη Νοημοσύνη και Ρομποτική

29 Οκτ 2013 (πριν από 3 χρόνια και 10 μήνες)

103 εμφανίσεις





Gravitant Advanced Analytics

Copyright © 2010 Gravitant, Inc.


W H I T E P A P E R





Advanced
Analytics for
Intelligent
Choices

February 5

2010

Applying the “glue” between the business and IT
Operations Management and leveraging “Operation
s

Research” algorithms to deliver decision management for
the Agile CIO.

Applying Operation
s

Research Algorithms
for CIO Decision
Making


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
2



Executive Summary

In today’s world there is little to no separation between business strategy and operations
from IT strategy and operations. Companies must focus on the value in their operations,
exploit opportunities and act quickly to seize competitive advantage. There h
as never been
a time where so much hangs on the ability of the CIO and IT management staff to be
“Agile”.

CIO’s must be aligned and prioritized for the needs of the business. They must
ensure that they are optimizing service delivery with the highest qual
ity, in the most
efficient and timely way, and delivering upon the highest levels of customer satisfaction.


More than ever they are being held to the highest levels of accountability and pressure to
deliver and perform for the business and manage constant change.
CIO’s must re
-
think IT
and focus on value delivery to strategic busin
ess objectives.

CIO’s and their team want to
know:

1)

How well they are prioritized, aligned and performing for the business.

2)

If they are under
-
performing, why?

3)

The
y

want to know where to optimize existing performance delivery.

4)

They want to know how to optim
ize for change.



The CIO needs Advanced Analytics for the business!

Why is Advanced Analytics Required Today?

Improving Performance through Business Model Innovation

For the past several decades, organizations have focused on increasing profitability
thr
ough automation, process re
-
engineering and new technology. Many of these initiatives
have been successful, and companies have enjoyed improved financial results. But the
quick wins are over, and the competitive landscape has leveled as companies adopt sim
ilar
strategies. The challenge today is building sustainable competitive advantages for the
future, especially when the macro
-
economy is changing at an ever
-
growing rate.

Business leaders and industry pundits are in agreement that business model innovation

is
the key to developing sustainable competitive advantages. IBM’s 2008 Global Study of
1,000 public and private sector leaders reported that “nearly all CEOs are adapting their
business models


two
-
thirds are implementing extensive innovations.” Executi
ves
clearly recognize that successful companies of the future must react quickly to market
forces.

Fortunately there’s a wealth of information to help executives make their decisions about
business designs. In fact, there’s so much data in enterprise syst
ems that the pendulum has
swung in the direction of information overload. Business leaders want to make bold
decisions, but they’re finding themselves weighed down with disparate information. It has
Business model

is t
he
network of customers,
vendors, and
enterprise business
components designed
to
create

value
. Most
large organizations
have multiple business
models.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
3


become increasingly difficult to identify and take advant
age of relevant, real
-
time
information from a multitude of sources.

To gain confidence in their decisions, business leaders need to recognize and address the
following challenges:



Business models have become exponentially more complex
. The Internet
has
opened the doors to a global marketplace. It’s now possible to collaborate
with organizations around the world, which often results in a web of
nonstandard, but critical, relationships.



The human mind can’t effectively map all the interconnected
relationships within

a large organization.

Without a blueprint of a
company’s business model and its operations model (in other words
--

the
entire value chain), it’s impossible to accurately predict the full effect of major
changes to an organization.



Timely, accurate and m
eaningful data is required for decisions
. Business
decisions are based on data, and incomplete or out
-
of
-
date data will lead to
bad decisions. The trick is identifying meaningful data that can be turned into
actionable information through analytics. Accord
ing to Forrester, “digital data
volumes


both structured and unstructured


are growing by 30% per year
and will be approaching 1 zettabyte by 2010.”
1

That’s a one with 21 zeros.
Compounding the problem of sher massive quantities of data is the fact that
data is usually siloe

d in different IT systems. Regardless, access to the right
data in real
-
time is absolutely required for intelligent decision
-
making.

Although these challenges are substantial, advanced analytics provides a
scientific
approach and

solu
tion

to help with business model innovation, while effectively taking
advantage of the right enterprise data.



Advanced analytics is the foremost solution to aid companies with
business model innovation.
Only advanced analytical solutions that are
grounded
in a blueprint of an enterprise business model can help executives
with strategic decision
-
making. Sophisticated analytic techniques that
incorporate the business design can allow business leaders to test scenarios,
optimize transformative plans and predic
t outcomes. In other words, advanced
analytics can effectively remove risk from major decisions that have long
-
term
implications.

Gravitant’s Unique Approach to Advanced Analytics

Gravitant is on the forefront of applying complex Operations Research (OR) t
echniques to
business problems; techniques that have long been used to optimize airline patterns and
military movement. Our Advanced Analytics

includes basic business analytics, such as
linear regression and descriptive statistics. But it also includes mor
e recondite techniques
for business management, including demand forecasting, impact analysis, root
-
cause
analysis, simulation and optimization.

Operations Research

applies
mathematical
modeling
,
statistics

and
algorithms

to
arrive at optimal
solutions to complex
problems. It uses
many solution
techniques, including
stochastic
optimization, queuing
theory and Monte
Carlo simulation.



Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
4


Case Study
: Gravitant is using advanced analytics to model the
impact of hurricanes on the State of Texas Hea
lth and Human
Services (HHS) agency after Hurricane Ike pummeled the Texas
coast in 2008. The model predicts the effect of future natural
disasters on the agency’s call centers, IT operations and
workforce. It quickly predicts bottlenecks in the systems an
d the
impact on cost, customer service and utilization.

This flexible model could be adapted for any health agency in the
U.S. to predict outcomes of catastrophic events, such as viral
outbreaks, floods or hurricanes.

Compared to data mining and business

intelligence programs, Gravitant’s Advanced
Analytics provides a broader context of insight and interpretation for users. Our goal is to
equip companies with real
-
time business information that identifies the consequences of
any given business decision


whether it affects suppliers, partners or customers. Our
Advanced Analytics are stunningly insightful, because the organization’s business model is
used as the basis for the results. With that context, business leaders can quickly see
enterprise
-
wide impac
t and confidently develop actionable plans.

Gravitant’s Advanced Analytics

Re
al
-
world application of advanced analytics to
IT

management is a relatively new and
powerful concept. Gravitant
is a pioneer in applying OR techniques to business models
with its
IT Management Solutions on a BusinessMatrix Platform, which is the
revolutionary Enter
prise Program & Portfolio Managem
ent (EPPM) platform

Figure
1

s
hows an
overview

of BusinessMatrix.

Gravitant’s IT Management is
a
flagship

solution for innovating business models,

linking
strategy with execution and managing change. Our web
-
based
application allows

IT
leaders’ to align and control IT service delivery
, reduce cost, minimize risk, control
transformations and drive business efficiency.


At the core of the solution is a
Performance Management

system that stores the enterprise
business and

operations
meta
model. Th
is model

create
s

a shared reference point for the
organization. Our Advanced Analytics wraps around this
enterprise meta

model for
seamless analysis and impeccable foresight for testing and optimizing corporate
strategies.


Regression

is a
method of deriving
relationships between
variables and
outcomes. It uses a
method of minimizing
the error between the
actual value and the
predicted value.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
5



Figure
1
: BusinessMatrix Solution

Traditional Methods of Decision Support

In developing BusinessMatrix Advanced Analytics, the Gravitant team examined how the
majority of companies used analytics for traditional methods of decision support.
Of all
the tools on the market, regression is the most widely used
to analyze input vari
ables and
predict likely performance outcomes directly tied to those variables. (See
Figure
2
.)

Figure
2
: Traditional Busi
ness Analytics

Regression has been widely adopted, because the tools are readily available and its
application is deceptively straightforward. Among the numerous benefits, regression can
predict outcomes directly associated with changes to input variables. Examples of
co
mmon input variables are demand, resource capacity and resource capability. As
business users change the input variables, they can quickly
obtain intermediate predictive
results, such as approximations for future demand and resource utilization. These
est
imates are relatively easy to obtain, because they do not require extensive data
organization.

However, r
egression can’t predict the
downstream

impact of changes, which is particularly
important in business situations. Companies want to know the effect of

decisions through
Operations

model

is
t
he process
es
,

people,

resources and assets
within an enterprise
that
support the value
created by that
organization.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
6


their entire financial and operational systems. Regression alone can’t provide this. Also,
regression is often criticized for three other reasons:

1.

It relies heavily on the underlying assumptions,

2.

Users often have to define correlations

that may not exist, and

3.

It requires an overwhelming volume of consistent data to generate accurate
results.

BusinessMatrix Advanced Analytics Addresses Pitfalls of Traditional
Methods

Although regression has important applications in decision support,
the Gravitant team
realized that users need more sophisticated tools to stay competitive in the 21st century.
The objective of Gravitant’s Advanced Analytics is to provide accurate, intermediate,
downstream and enterprise
-
wide business intelligence. It is
a breakthrough approach to
addressing the increasing complexity and need for decision support.

BusinessMatrix Advanced Analytics offers three critical innovations:

1. The organization of real
-
time performance data to create a blueprint of the
enterprise
business and

operations
meta
model
s
. This
enterprise
business
meta model provides context and is stored in BusinessMatrix’s
Performance
Management

system. Without the meta model, advanced analytics cannot
truly show the enterprise
-
wide impact of decisions.


2. The use of OR techniques to create advanced analytics that provides
unparalleled insight. These techniques provide more accurate results on
downstream effects, like financial performance and customer satisfaction.
(See
Figure
3
.
)

3. BusinessMatrix Advanced Analytics requires significantly less data to
predict more accurate outcomes than regression.

Instead of using many
assumptions and manual inferences in the data, our
enterprise

meta model
structures those assumptions into conc
rete concepts and creates the flow of
information/data in a well
-
defined system.

Given that system definition, less
historical operational data is needed to analyze options for changes to the
system.

Moreover, all assumptions and predictions are improved o
ver time by a feedback
mechanism from real
-
time data.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
7


Figure
3
: BusinessMatrix Advanced Analytics

BusinessMatrix Advanced Analytics: How it Works

Digging deeper into the mechanics of BusinessMatrix Advanced Analytics, business u
sers
will find an extensive combination of
Intelligence Modules
, Gravitant

Models

and OR Tools.
To understand how BusinessMatrix Advanced Analytics works, it’s important to
understand the hierarchy and interactions of these components. (See
Figure
4
.)


Figure
4
: BusinessMatrix Advanced Analytics

Starting at the very core of BusinessMatrix Advanced Analytics shown in
Figure
4

is the
enterprise meta

model that represents the business model and operations model. It is

Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
8


stored in BusinessMatrix’s
Performance Management

system (
Figure
1
). The
enterprise
meta

model is created using Gravitant’s patent
-
pending methodology, which identifies and
organizes data into a flexible framework that is always pulling real
-
time

data from
enterprise IT systems. The data collected to form the enterprise meta model includes the
key performance indicators (KPIs) of the enterprise. Most organizations are already
tracking the majority of these KPIs, but often new KPIs are uncovered th
rough the process
of defining the business meta model.

After the KPIs are identified and placed in analytical matrices, algorithms are developed
that form the linkages between the matrices. These linkages create the flexible framework
that represents the
enterprise
meta model. When this process is complete, the enterprise
meta model is well defined. Business users know the important KPIs to be monitoring.
They know what variables are under their control, such as service level agreements with
vendors. They
also know what variables are out of their control, such as demand or
unexpected events.

Defining the
enterprise
meta model provides the parameters for the hierarchy shown in
Figure 4. Using these parameters, the components of Gravitant’s Advanced Analytics are
then utilized to achieve desired results.

The top cylinder of
Figure
4

contains the
Intelligence Modules
. This component is
responsible for two functions. First, it determines the best
Gravitant Models and OR T
ools
to solve the business user’s problem. The
re is a one
-
to
-
many relationship between both
the
Gravitant M
od
el
s and
OR T
ools, so utilizing the right combination is important when
calculating results. Second, the
Intelligence Module

feeds the end results to
BusinessMatrix’s web
-
based solution for simp
le visualization of the results.

The middle cylinder in
Figure
4

contains Gravitant’s patent
-
pending intellectual property.
These sophisticated mod
els

are responsible for doing the complex calculations and
leveraging the OR tools
for simulation and optimization
. Gravitant

Models

are specifically
designed to incorporate real
-
time data from the
enterprise
meta model for unprecedented
context and accuracy
of results. The mod
el
s automatically determine which OR techniques
are most appropriate given the nature of the problem, the goals for improvement, and the
constraints on the business services.

The bottom cylinder of
Figure
4

consists of the OR tools used by the Gravitant
Analytics
mod
u
l
e
s. These OR tools are particularly suited for business

model optimization

because
they are distinguished in the mathematics field for examining entire systems, rather than
concentrating only on specific elements.

Details on the
Intelligence Modules

The
Intelligence Modules

are industry
-
specific and have built
-
in domain knowled
ge for
running the analyses and displaying the results in a manner that makes
sense to the user.
These
module
s

perform
active

and predictive analyses leveraging the
enterprise

meta
model, the right Gravitant
Models

and OR
T
ools for the situation. The
Inte
lligence Modules

include the following components.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
9


Active

Analytics Module

Active

Analytics
identifies process bottlenecks. Business users can quickly identify and
focus their attention on the 20% of operations that are causing 80% of the issues.



Descriptive Statistics:

Describes the data by providing basic measures such as
mean, mode, variance, quantiles, etc. Uses basic statistical formulae.



Inferential Statistics:

Identifies 20% of the problem areas that cause 80% of
the issues. Uses descrip
tive statistics along with pareto analysis.

Case Study
: Gravitant’s
active

analytics is used to monitor the
performance of the Family and Social Services Administration
(FSSA) in the state of Indiana. In 2008, the Governor and FSSA
promoted a new citizen program that instantly gained traction in
the community. Citizen demand for

the program quickly and
unexpectedly backlogged the FSSA’s eligibility system, which is
made up of a complex network of outsourced partners. The state
used Gravitant’s descriptive and inferential statistics to help
identify the 20% of work queues


whethe
r internal or external
--

that were causing 80% of the backlog. The state acted on the
information and added more staff to the right processes to
effectively reduce the backlog.

Gravitant’s
Active

Analytics is particularly effective in this
situation, bec
ause it’s able to uncover root causes in large, co
-
dependent systems using the
enterprise

meta model.

Predictive Analytics Module

Predictive Analytics
helps align resources to optimize financial and operational
performance. It is most often used to test n
ew strategies and rule out plans that are likely
to fail if implemented. When testing scenarios, predictive analytics can identify future
bottlenecks. It can suggest resolutions and show the enterprise
-
wide impact.



Forecasting:

Predicts demand from histo
rical data, including the effects of
seasonality and other events. Uses stepwise linear regression.



Impact Analysis:

Estimates the impact of demand on other performance
metrics. Uses linear regression.



Root Cause Analysis

o

Diagnostics:

Identifies the roo
t cause of an issue. Uses the analytics
framework.

o

Prognostics:

Estimates the impact of the root cause issue on other
performance metrics. Uses the analytics framework and balanced
scorecard performance management framework.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
10




Bottleneck Resolution

o

Scenari
o Analysis/Simulation:

Provides the capability to simulate the
impact of decisions on performance metrics. Uses the analytics
framework, the balanced scorecard framework and simulation
algorithms.

o

Optimization:

Identifies the optimal decision subject to
the pre
-
defined constraints on the performance metrics. Uses the analytics
framework, the balanced scorecard framework and optimization
algorithms.

Case Study
: Gravitant’s predictive analytics is capable of
accurately predicting how policy changes, natura
l disasters and
seasonality will affect Health and Human Service state
-
agencies.
The model has been developed using domain expertise, real
-
world benchmarks and advanced analytics.

In the State of Texas, our predictive analytics tool can identify
bottleneck
s in the case lifecycle. It also predicts the impact of this
bottleneck on performance criteria such as cost, timeliness and
resource utilization.

The tool then allows decision makers to virtually simulate the
effect of resource changes, whether they are w
orkforce, processes
or technology. While many options could reduce the impact of the
bottleneck, the tool has built
-
in mathematical optimization
algorithms that identify the best solution that doesn’t violate any
system constraints.

As a result, decision m
akers can evaluate solution scenarios and
make balanced decisions with the costs and risks scientifically
predicted.

Transformation Analytics Module

Transformation Analytics identifies

optimal path and timing of initiatives to transform an
organization to
achieve its strategic goals. Transformation Analytics takes into
consideration all the constraints on a system, including financial, workforce, processes and
technology.



Initiatives Generation:

Proposes a set of feasible initiatives for transformation.
U
ses the analytics framework, the balanced scorecard framework and domain
knowledge.



Roadmapping
: Identifies the optimal transformation decisions, subject to
constraints on performance metrics. Uses the analytics framework, the balanced
scorecard framework

and random selection optimization.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
11


Case Study
: The Department of Information Resources (DIR) in
the state of Texas has the Texas
-

Agency Network (TEX
-
AN) to
source and manage all telecommunications services for state
agencies. As TEX
-
AN grows and evolve
s, DIR is looking to
Gravitant’s transformation analytics to identify the optimal
workforce that will maintain customer satisfaction while
meeting budgetary constraints. Operations managers can use
such a tool to maintain operational performance while enab
ling
changes to the business model.

Gravitant Models

Gravitant models are developed by our Operations Research scientists for the explicit
purpose of
IT & B
usiness management. The key differentiator for the models is their usage
of the
enterprise meta

model in all the calculations. Without that context, the analytics
would not be able to effectively predict and optimize the enterprise
-
wide impact of
strategic decisions. Gravitant has developed the following models.



G_Stat:
Helps discover unexpected p
henomena in monitored data. Correlates
multiple datasets to identify possible relationships. Provides pareto analys
e
s to
identify potential issues.



G_Forecast
:
Forecasts future demand based on historic trends and seasonality.
Forecasts independent and d
ependent variables, which quickly shows
secondary effects of expected and unexpected changes. Provides confidence
interval
s

on that forecast for both independent and dependent variables
.



G_Business
:
Helps align partners along with internal operations to s
atisfy a set
of customers. Allows users to run scenarios and analyze the tradeoffs between
vendor costs and customer satisfaction.

Identifies the optimal value chain of
customer segments, channels and vendors. Automatically recommends the best
strategies t
o achieve desired financial and customer satisfaction goals by
running combinations and determining the best mix of vendors and customers.



G_Operations
:
Estimates the impact of resource capacity and capability on
operational performance. Quickly tells th
e user how additional workforce and
technology (capacity) impact operational goals. Tells users what capabilities are
required, such as worker skill sets or IT service availability. Helps users
understand the tradeoffs between cost, resource utilization an
d customer
satisfaction. Identifies the optimal resource capacity and capability to achieve
operational goals. Automatically tests combinations of workforce and
technology in terms of the capability and capacity to determine the best
combination based on u
ser
-
defined constraints.



G_Transform
:


Predicts the success of initiatives based on performance and
cost. Tells users which initiatives will actually improve business outcomes.
Helps users allocate more resources to critical initiatives. Automatically

Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
12


schedules projects for optimal outcome w
ithout violating dependency
constraints.



G_Finance:

Helps
ensure financial stability by efficient spending. Allows users
to reduce the net present worth of investments by paying them off at the
optimal time. Automatically allocates remaining budget to ven
tures that are core
to the business.

The Intelligence Modules use the Gravitant Models along with the real
-
time
Enterprise
Meta

Model data in the manner shown in
Table
1
.

Table
1
: Intelligence Module to Gravitant Model Relationships

Module

Models Required


G_Stat

G_Forecast

G_Business

G_Operations


G_Transform

Active
Analytics



Descriptive statistics








Inferential statistics







Predictive Analytics



Forecasting








Impact analysis









Root cause analysis



Diagnostics








Prognostics








Bottleneck resolution



Scenario
analysis/Simulation










Optimization








Transformation Analytics



Initiatives generation








Roadmapping








Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
13


OR Tools

A large number of Operations Research (OR) tools are used to support each of the
Gravitant Models. The classification
of tools can be seen in

Table
2
.
Table
3

shows the
mapping bet
ween tools and models.

Table
2
: List of OR Tools

Tool Name

Tool Description

Probability & Statistics




Regression

Tool for identifying the relationship between two or
more sets of data using the least squares error technique


Bootstrapping

Tool for generating a larger sample data set by strapping
a distribution around the original data set and then
generating instances from this distribution


Queuing

Tool for identifying the number of resources required for
a given arrival
rate of demand and service rate of the
resources

Simulation




Event
-
based simulation

Tool for evaluating complex scenarios where randomly
generated events impact the outcome


Monte Carlo simulation

Tool for solving complex problems by generating a
large
number of random instances and observing the most
common outcome

Optimization




Linear programming

Tool for optimizing an objective subject to constraints,
where the constraints are linear, decision variables are
linear and continuous, and
parameters are deterministic


Integer programming

Tool for optimizing an objective subject to constraints,
where the constraints are linear, decision variables are
linear but not necessarily continuous, and parameters are
deterministic


Nonlinear programming

Tool for optimizing an objective subject to constraints,
where the constraints are not necessarily linear, decision
variables are not necessarily linear, and parameters are
deterministic


Stochastic programming

Tool for optimizing a
n objective subject to constraints,
where the parameters are not all deterministic


Tools under probability and statistics are used to approximate values that cannot be
calculated directly. Under certain conditions, these approximations may not be
theoret
ically possible, in which case one would resort to simulation tools.

However, if it is possible to approximate values using probability and statistics, then
optimization tools can be used to identify the best solution among a feasible set of options.
Otherwise, simulation
-
based optimization would have to be implemented to find the best
solution. In some cases it may even be possible to use specialized algorithms that have
already been developed in operations research literature.


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
14


Table
3
: Model/Tool Relationships


OR Tools Required

Model


Regression


Bootstrapping


Queuing


Linear programming


Integer programming


Nonlinear programming


Stochastic programming

Event
-
based simulation

Monte Carlo Simulation

G_Stat










G_Forecast












G_Business












G_Operations












G_Transform












How BusinessMatrix Advanced Analytics is Deployed

There is a common process and methodology to successfully deploy BusinessMatrix for
the engagement.

1.

Setup the
Enterprise Meta Model.

2.

Based on the analytics module of interest, setup one or more of the
following:

o

Business Model,

o

Operations model, or

o

Transformation
model.

3.

Calibrate models based on monitored metrics and tracked initiatives.

4.

Test analytics using
historical data.

Figure 5:
A
d
v
a
n
c
e
d

A
n
a
l
y
t
i
c
s

Deploy
ment Flow Chart


Gravitant Advanced Analytics



Copyright ©
2010
Gravitant,
Inc.

pg.
15


Conclusion

Gravitant IT Management software can deliver proven business agility for the CIO and
his/her staff. Business Matrix provides the common agreement, understanding, language
between the business and IT. That common baseline of understanding can be leveraged t
o
deliver Advanced Analytics to measure performance of IT Service delivery in context of the
associated business priority and expectations. With Business Matrix the CIO and his/her
staff can leverage advanced OR methods to provide predictive analytics on f
orecasting and
simulation with context to the business for optimal decisions. Finally with Business Matrix
the CIO and his/her staff can leverage OR methods
utilize

calculate complex optimization
models to determine the optimal IT service delivery based on

input constraints. With
Gravitant, CIO’s now can answer:

1)

How well is IT performing for the business?

2)

When things are under performing, why?

3)

What is the optimal
IT service model to deliver to the needs of the business?

About Gravitant

Gravitant provides n
ext
-
generation Enterprise Program &

Portfolio Management (EPPM)
solutions for IT

management. These solutions are built on a service
-
based

business/IT
value chain. With Gravitant’s solution, CIO &

IT executives can now

effectively make
investment decisions
based on an analysis of business service,

performance, risk, IT
outsourcing and cost trade
-
offs.