Interaction analysis

from Wikipedia, the free encyclopedia

The term “ interaction analysis ” or “ cross-impact analysis ” refers to a forecasting technique that attempts to represent and analyze relationships ( cross impact ) between different events that may occur in the future , and to take their mutual effects into account. We know from experience that most events or developments are in some way related to other events and developments (" correlation "). Many other forecasting techniques (such as the Delphi method ) can only look at one delimited problem. The links between individual events are not taken into account. Cross-impact analysis fills this gap. It is therefore u. a. used in the scenario technique .

The probability that a certain event will occur is directly influenced by the occurrence or non-occurrence of another event. The cross-impact analysis makes it possible to determine the probability of an event occurring depending on other events. It was developed in 1966 by Theodore Gordon and Olaf Helmer and resulted from the simple question: Can predictions be based on how future events will influence each other? The first application of the cross-impact method was in connection with a game ("Future") that Gordon and Helmer developed for the Kaiser Aluminum and Chemical Company .

method

1. Identifying the events : The first step in a cross-impact analysis is to identify the events to be considered. This step is critical to success. On the one hand, all relevant developments that have not been taken into account but still have an influence can falsify the result. On the other hand, too precise an analysis, taking into account every conceivable event, can complicate the study unnecessarily. Since the number of interactions between the different pairs is equal to -n (n = number of events), usually between 10 and 40 events are taken into account. This initial list of events is usually done by summarizing existing data and interviewing experts .

2. Estimate probabilities of occurrence : When estimating the probabilities of occurrence for each event, each event is considered independently / in isolation, i. H. without taking into account possible influences from other developments.

3. Calculate conditional probabilities : Then the conditional probabilities are determined. The following question is asked for each pair of events: If event m occurs, what is the new probability of occurrence for event n? This creates the cross-impact matrix :

When this
event occurs ...		 ... the probability of occurrence changes from ...
Initial
likely
friendliness		 1 2 3 4th
Event 1 0.20 X 0.90 0.50 0.15
Event 2 0.70 0.35 X 0.20 0.30
Event 3 0.35 0.10 0.40 X 0.05
Event 4 0.10 0.15 0.50 0.60 X

The interpretation for the respective event pairs can now take place: Event 2 has a probability of 0.70 - considered in isolation. However, if event 1 occurs, the probability that event 2 will also occur increases to 0.90. Likewise, the - viewed in isolation - probability that event 3 occurs is 0.35. But if event 2 also occurs, the probability of event 3 occurring is reduced to 0.20.

4. Sensitivity analysis : After the cross-impact matrix has been completed, several test runs are carried out with a computer program in order to better coordinate the matrix. Events are selected at random, and the occurrence or non-occurrence and the resulting influences on all events are calculated.

rating

A critical examination reveals some weaknesses of the cross-impact analysis:

  • the selection and assessment of the relevant factors is subjective
  • the analysis is based on data pairs - in the real world, however, several developments can influence an event at the same time
  • Collecting and evaluating the data can be very time-consuming - for example, 870 influences have to be calculated for the evaluation of 30 possible events

Nevertheless, it is precisely this detailed study of various influencing factors and their effects that is one of the greatest advantages of cross impact analysis. It can provide decisive food for thought for alternative approaches or show new ways.

See also

Scenario technique

literature

  • Luis F. Alarcón, David B. Ashley: Project Management decision making using cross impact analysis , International Journal of Project Management Vol. 16, No. 3, pp. 142-152,1998
  • U. Asan, CE Bozdag, S. Polat: A fuzzy approach to qualitative cross impact analysis , Omega, Vol. 32, No. 6, pp. 443-458, 2004
  • Enzer, Selwyn: Cross Impact Techniques in Technology Assessment , Futures, Vol. 4, No. 1, 1972
  • TJ Gordon: Cross-Impact Method , 1994
  • TJ Gordon and H. Hayward: Initial Experiments with the Cross-Impact Method of Forecasting , Futures, Vol. 1, No. 2, pp. 101-116, 1968
  • William G. Sullivan, Ph.D .: A Cross-Impact Analysis of the solar space heating and cooling industry , Industrial Management - July-August, 1978
  • Fabiana Scapolo; Ian Miles: " Eliciting experts' knowledge: a comparison of two methods ", Technological Forecasting and Social Change 73, p. 679-704, 2006

Web links

See also