Cross impact-analysis

CROSS IMPACT ANALYSIS
( CONCEPT & PRACTICES)
Presented by
JHA PRAVINKUMAR ( I-15-18-7)
SHINGALA SANKET ( I-15-18-17)
TRIPATHI MANISH ( I-15-18-19)
UKKOJI UTTAM ( I-15-18-20)
BOURA SANJAY SINGH ( I-15-18-24)
Thakur Institute of Management Studies
&
Research
(Sunday, 25 February 2018)
1

WHAT IS CIA?
• A family of techniques designed to evaluate
changes in the probability of the occurrence of a
given set of events consequent on the actual
occurrence of one of them
• An analytical approach to the probabilities of an
item in a forecasted set
• Its probabilities can be adjusted in view of
judgments concerning potential interactions
among the forecasted items
2

WHAT IS CIA?
• Interrelationship between events and
developments is called "cross-impact"
• The method tries to answer the question: can
forecasting be based on perceptions about how
future events may interact?
• Originally developed by Theodore Gordon and Olaf
Helmer in 1966
3

HISTORY OF THE CIA
• The origin of cross-impact analysis was the problem
that Delphi panelists were sometimes asked to make
forecasts about individual events, when other events
in the same Delphi could affect these events
• Thus, it was recognised that there was a need take
these cross impacts of one event on another into
account
• While cross-impact analysis was initially associated
with the Delphi method, its use is not restricted to
Delphi forecasts
• In fact, cross impact models can stand alone as a
method of futures research, or can be integrated with
other method(s) to form powerful forecasting tools
5

HISTORY OF THE CIA
• The Central Intelligence Agency (CIA) became
interested in the methodology in the late 1960s
and early 1970s as an analytic technique for
predicting how different factors and variables
would impact future decisions
• In the mid-1970s, futurists began to use the
methodology in larger numbers as a means to
predict the probability of specific events and
determine how related events impacted one
another
• By 2006, Cross Impact Analysis matured into a
number of related methodologies with uses for
businesses and communities as well as futurists
and intelligence analysts 6

EARLY EXPLORATION PHASE
In initial attempts to collect judgments about
the quantification of these interactions,
researchers recognized that interactions
among events constitute a powerful way to
examine perceptions about the future.
8

PROBABILISTIC PHASE
How can the conditional probability questions be asked?
When an expert is asked to provide judgment about the
probability of an event, does he or she:
(a) include the possibility of the cross impacts, a priori; or
(b) are the events seen as standing alone?
Given that each event has an initial probability of one sort or
the other, and, given the possible occurrence or
nonoccurrence of an event, the conditional probabilities
provided by expert judgment must meet certain coherent
limits. These limits can be calculated.
9

SYNTHESIS PHASE
Cross impact can stand alone as a method of futures
research or can be integrated with other methods to
form powerful tools. When integrated, cross impact
allows the introduction of perceptions about the
future into otherwise deterministic methods . In
addition, various methods of collecting judgments
(e.g., Delphi, mailed questionnaires, interviews, etc.)
have been used in conjunction with cross impact to
simplify the data gathering process.
10

APPLICATION PHASE
In recent years, the work on cross impact has shifted from
"pure“ methodological development to applications.
Questions about the method remain, of course: how best to
ask questions about conditional probabilities; is the method
really convergent; how to handle noncoherent input from
experts; how to integrate with other methods? But there is
no doubt that cross-impact questions help illuminate
perceptions about hidden causalities and feedback loops in
pathways to the future.
11

Cross Impact Analysis Architecture
13

#1 Choice of the issue and selecting experts
• The purpose of a cross-impact exercise is primarily to gain
more insight into future developments.
• Future developments may be defined as the result of
interactions between trends, events and the actions of
societal actors, thus the collection of information on the
historical background of the selected issue is important to
better focus on a limited number of aspects which can play
a role in the characterisation of future developments of
the selected issue.
• During this step a preliminary list of events related to the
issue could be formulated.
• The experts chosen should be familiar with the issue under
study and they should have some capacity to envisage
future developments.
14

• However, as for all other techniques that rely on
eliciting expert opinion, there is the problem of
avoiding bias in the group of experts.
• It is not obvious how to define 'relevant expertise'
when complex technological, social and political issues
are involved. There are no clear guidelines, either on
whether it is better to have a panel of experts which
involves experts from various sub-disciplines of the
subject considered, or if it is better to have experts
that are highly specialised or generalists with a broad
view.
15

• Experts are normally asked to do the following :
1. Appraise the simple probability of a hypothesis
occurring by means of a scale from 1 (very low
probability) to 5 (highly probable)
•2. Appraise the conditional probability of a
hypothesis if the others occur or not.
• Given these questions, the experts have to show
the level of implicit coherence in their reasoning.
16

#2 Final selection and definition of the events
• This step could be crucial to the successful implementation
of the method, in fact any influence not included in the
events' set will be completely excluded from the study.
• On the other hand, the inclusion of irrelevant events can
complicate the final analysis of the results unnecessarily.
• The final list of events should be as clear as possible,
definitions and wording must be carefully checked and
defined.
• The selection of events to be included in the final list can
cover both the occurrence and non-occurrence of events
• Then, the events considered can be totally independent or
connected in some way. The final list of events can also be
compiled with the support of experts on the selected
issue, or can stem from other methods used to collect
opinions, such as the Delphi method. 17

#2 Final selection and definition of the events
• Since the number of event pair interactions to be
considered is equal to n2 - n (where n is the number
of events), the number of interactions to be
considered increases rapidly as the number of events
increases.
• Most studies include between 10 and 40 events.
18

#3 Design of the probability scale and
definition of the time horizon
• The definition of a probability scale is needed to
translate qualitative appreciation from the experts on
the degree of occurrence into probabilities.
• The meaning of the scale must be clearly defined, so
not to occur in misunderstanding which could distort
the forecast.
• In general, the probability scale for cross-impact
methods usually goes from 0 (impossible event) to 1
(almost certain event).
19

#3 Design of the probability scale and
definition of the time horizon
• This step also involves determining the time horizon of
the forecast. In the context of Foresight the main
objective is to try to think ahead in the long term.
• Therefore, in Foresight the short term is considered to
range from the present to five years from now; the
medium term from five to ten years; and the long
term from twenty to fifty years. In fact one of the
main differences between Foresight and planning is
the temporal dimension. The time horizon to be
considered in a cross-impact analysis must be stated
explicitly.
20

#4 Estimating probabilities
• In this step the initial probability of the occurrence of
each event is estimated.
• Then, conditional probabilities in a cross impact
matrix are estimated in response to the following
question: 'If event x occurs, what is the new
probability of event j's occurring?' The entire cross-
impact matrix is completed by asking this question for
each combination of occurring event and impacted
event.
• The specific cross-impact method mainly described
here is the SMIC (Cross Impact Systems and Matrices)
method, which was developed in France in 1974 by
Duperrin and Godet.
21

• The SMIC is designed to enable experts' estimates to be
checked for consistency.
• The SMIC method invites the experts to answer a grid the
following questions:
1. the probability of occurrence of each single event at a
given time-horizon
2. the conditional probabilities of the separate event taken in
pairs at a given time-horizon:
P(i/j) probability of i if j occurs
P(i/not j) probability of i if j does not occur.
• Once the results are collected, they are entered on the
computer and the program is run.
22

• The calculation for a range of conditional probabilities
that will satisfy this consistency requirement is easy.
• The initial probability of an event can be expressed as
follows:
P(l) = P(2) x P(1/2) + P(2c) x P(l/2c) (1)
where:
P(l) = probability that event I will occur;
P(2) = probability that event 2 will occur;
P(1/2) = probability of event 1 given the occurrence of event
2;
P(2c) = probability that event 2 will not occur; and
P(1/2c) = probability of event 1 given the nonoccurrence of
event 2.
23

• The limits on the new probability of event 1 given the
occurrence of event 2 are:
{P(1) - 1 + P(2)}/P(2) <= P(1/1) <= P(1)/P(2)
24

• Once the cross-impact matrix has been estimated, a
computer program is used to perform a calibration run of
the matrix.
• A run of the matrix consists of randomly selecting an event
for testing, comparing its probability with a random
number to decide its occurrence or nonoccurrence, and
calculating the impacts on all the other events due to the
occurrence or nonoccurrence of the selected event.
• Impacts are normally calculated using odds ratios. To apply
the odds ratio technique, the initial and conditional
probabilities of the events are converted to odds, using
the following relationship:
Odds = Probability / 1 - Probability
25

• Once the odds ratios have been determined, the
calculations proceed as follows:
1. An event is selected at random from the event set.
2. A random number between 0.0 and 1.0 is selected. If the
random number is less than the probability of the event being
tested, the event is said to occur.
If the random number is greater than the event probability,
the event does not occur.
3. If the event (event j) occurs, the odds of the other events
occurring are adjusted as follows:
New odds of event i =
(initial odds of event i) x (occurrence odds ratio of event j on
event i)
26

• If the event does not occur, the same calculations
are made but the nonoccurrence odds ratios are used.
4. Steps 1, 2, and 3 are repeated until all the events have
been tested for occurrence.
5. Steps 1 through 4 (which represent one play of the
matrix) are repeated a large number of times.
6. The frequency of occurrence of each event for all runs
of the cross- impact matrix determines the new
probability of that event.
27

#5 Generation of scenarios
• The outcome of applying a cross-impact model is a
production of scenarios.
• Regardless of how the issue of assigning probabilities
is resolved in specific cross-impact models, the usual
procedure is to carry out a Monte Carlo simulation
• Each run of the model produces a synthetic future
history, or scenario, which includes the occurrence of
some events and the non-occurrence of others. The
model is thus run enough times (i.e. approximately
100), so that the collection of output scenarios
represents a statistically valid sample of the possible
scenarios which the model might produce.
28

• In a model with n events 2n possible scenarios
are generated, each differs from all the others in
the occurrence of at least one event
• It is worth noting that the number of runs
required increases exponentially with the
number of events. For example, if there are 10
events to be considered, there are 1024 possible
scenarios to estimate.
29

• On the basis of the specific cross-impact model
applied, the output scenarios attempt to generate
either the best scenario - in the sense of likelihood
of occurrence; or a set of statistically consistent
scenarios; or one or more plausible scenarios from
the total set.
• The SMIC method generates a cardinal sequence of
possible scenarios (from the most probable to the
least probable). This allows to circumscribe the area
of plausible future developments by retaining only
those which have a high-average probability of
occurrence. The list of scenarios generated by the
software need to be interpreted and described by
referring back to the original set of events. 30

• Once the cross-impact matrices are calculated, it is
possible to carry out a sensitivity analysis
• Sensitivity analysis consists of selecting an initial
probability estimate or a conditional probability
estimate, about which uncertainty exists.
• This judgment is changed and the matrix is run again.
• If significance differences appear between this run
and the original one, then it is apparent that the
judgment that was changed plays an important role. It
may be worthwhile to reconsider that particular
judgment
31

VARIATIONS OF CIA
• A simulation method, called Interax (1980), that
incorporated cross-impact concepts was developed by
Selwyn Enzer at the University of California (USA).
Ducos integrated Delphi and cross impact (1984)
• Bonnicksen at Texas A&M University (USA), in a
process called EZ-IMPACT, used the cross-impact
approach in a workshop gaming application to explore
policy options among contentious parties
• KSIM, a simulation technique developed by J. Kane
(1972) was based on expected interactions between
time-series variables rather than events; In this
approach, Kane treated all of the variables as a
percentage of their maximum value, and the cross
impacts were used to adjust the variables in each time
interval
32

VARIATIONS OF CIA
• Turoff generated scenarios from the cross-impact
matrix by assuming that events with probabilities
less than .5 did not occur and those with
probabilities equal to or greater than .5 did occur
(1972)
• Duval, Fontela, and Gabus at the Battelle Institute
in Geneva developed EXPLORSIM, a cross-
impact/scenario approach (1974)
• Duperrin and Gabus developed SMIC, a
crossimpact approach that asks experts to provide
initial, conditional occurrence, and conditional
nonoccurrence probabilities and to form scenarios
based on the cross-impact results (1974)
33

VARIATIONS OF CIA
• At The Futures Group, probabilistic systems dynamics
was a joining of systems dynamics and a time-
dependent version of cross impact, an approach first
explored by John Stover in simulating the economy of
Uruguay (1975)
• Scenario Management: Developed by Heiko Duin in
1995. Duin’s work has matured from scenario
management to long-term simulations using cross-
impact models. Duin is an innovator in using applying
the cross impact method to networked organisations.
His more recent work uses Virtual Organisation
Breeding Environments to help in enterprise network’s
strategic planning. The modeling framework aids in
generating strategic objectives and scenarios. 34

VARIATIONS OF CIA
• Delphi and Cross-Impact: Developed by Gilbert-François
Ducos in 1984. Ducos presented new concepts on the
Delphi method and what he called the Mini-Delphi.
• He then developed a technique to combine the Delphi
method to the cross impact analysis.
• The Delphi method consists of gathering a panel of experts
in order to have them individually create initial
probabilities along with arguments justifying their choices
on a set of events.
• All probabilities and arguments are then presented to the
panel after which each individual expert considers the new
information and alters his previous odds and arguments.
• After several rounds, individual probabilities should begin
to converge within the panel.
• Brent Vickers also demonstrated a computer assisted
approach to decision-making using the Delphi and cross
impact analysis methods. The technique was called
DELEWARE, a group decision support system (GDSS).
35

CHEMICAL INDUSTRY EXAMPLE
• Suppose a study of the future of the chemical industry was
in progress. In the course of the study, a list of important
future events is generated. One part of that list might
include the following events:
1. The use of plastics in transportation vehicles and
construction expands six fold from 1992.
2. Increased governmental intervention in the process of
innovation results from demands for consumer protection and
pollution control.
3. Chemical theory progresses to the point where much of
chemical research can be done through computer calculations
rather than actual experimentation.
4. The chemical industry expands into textiles and clothing
through the development of nonwoven synthetic fabric.
5. Chemical companies realize a declining return or rising
investment in conventional research.
36

• The first step in using these events in a cross-
impact analysis is to estimate initial probabilities
for the events. Experts, recognizing that all of
these events are possible and interact, might
provide the following probabilities:
37

• The next step is to estimate conditional
probabilities. In this step, a matrix is constructed.
Each cell of the matrix represents the answer to
the question, "If event x occurs, what is the new
probability of event y?“
38

• Since the influences of the events on each other were included
in the initial probability estimates, this judgment must now be
tested for consistency with the initial probabilities
• Using equation and the probabilities of events 1 and 2, we see
that the limits on the conditional probability of event 2, given
event 1, are 0.0 and 1.00. Thus, no problem is presented by the
judgment of 0.30 for the probability of event 2, given event 1.
{P(1) - 1 + P(2)}/P(2) <= P(1/1) <= P(1)/P(2)
• In a similar fashion, the entire matrix is completed. The next task
is specifying policy or sensitivity tests to be run with the matrix.
• In this case, we may wish to know the effect on the other events
if event 3 (use of computers for much chemical research) occurs.
• Thus, one test would be performed by assigning a probability of
1.0 to event 3 and rerunning the matrix.
• A second test might be performed to test the sensitivity of the
events to event 2 (increased governmental intervention in the
innovation process).
39

41
Thus, if event 2 were to occur, the principal consequence
would be an increase in the probability of event 5, from 20
percent to 29 percent.

WEAKNESSES
• The collection of data can be fatiguing and
tedious. A ten-by-ten matrix requires that 90
conditional probability judgments be made. A 40-
by-40 matrix requires that 1,560 judgments be
made
• This method assumes that, somehow and in some
applications, conditional probabilities are more
accurate than estimates of a priori probabilities;
this is unproved.
• As any other techniques based on eliciting experts'
knowledge, the method relies on the level of
expertise of respondents.
42

STRENGTHS
• Cross-impact methods forces attention into chains
of causality; a affects b; b affects c.
• Inserting a cross-impact matrix into another model
often adds power to that model by bringing into its
scope future external events that may, in the limit,
change the structure of the model
• Estimate dependency and interdependency among
events
• It can be used to clarify and increase knowledge
on future developments
• Use of groups of experts ensures a number of
opinions worth considering when calculating
probabilities of events
43

SAMPLES OF APPLICATIONS
• Aircraft construction
• World geopolitical evolution
• The nuclear industry
• Corporate activities
• Jobs
• European automobile industry
• Softwood lumber industry in Canada
• Economy of Uruguay
• Future of Hong Cong
44

Cross impact-analysis

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Cross impact-analysis

Similar to Cross impact-analysis (20)

More from A P

More from A P (20)

Recently uploaded

Recently uploaded (20)

Cross impact-analysis