dokumen.tips_pattern-planning-assistance-through-technical-evaluation-of-relevance-numbers

PATTERN
Normative forecasting
Relevance tree method
Goal-oriented forecasting method where one establishes a future need and recedes backwards to the present and to the technologies needed to achieve the objective of the future.

Basic form of relevance tree

Characteristics for forming the relevance tree:
There is hierarchy in the relevance tree
Branches represent goals and subgoals
All relevant subgoals for each goal have to be identified
Each branch must be well defined so that there are no overlaps

PATTERN
PATTERN has been used by the Honeywell Corporation for military, space and medical purposes.

PATTERN is based on:
Goal identification
Recognizing the relevance of set goals in relation to criteria (means ranking, e.g. setting goals priority)
Recognizing technological alternatives necessary for achieving the goal

Steps for PATTERN
Model description, recognizing the goals and hierarchy of the relevance tree
Recognize criteria
Determine relevance numbers – with participation of experts; selected exploratory, intuitive methods can be used
Data processing and final results – calculating relevance numbers, goals priority, ranking of technological alternatives

Basic terms
Goals A, B,C...j...N
Criteria α, β, …, x, …, v.
Levels 1, 2, 3...i...n
Criteria weights
Wα, Wβ, …, Wx, …, Wv.
Contribution marks of the goal j to criteria x - element weights
Sjα, Sjβ, …, Sjx, …, Sjv.
Based on the relevance tree primary matrix has to be made for each expert

Primary matrix

Basic terms
Based on the primary matrix the final primary matrix has to be calculated
The elements of final primary matrix are average values of responding elements in primary matrixes

Basic terms
A panel of experts can be asked to weight the importance of each criterium in relation to the others
The panel could be asked to weight the contribution of each element/goal to criteria – element weights

Conditions for primary and final primary matrix
Sum of criteria weights is 1
Sum of contribution marks of goals to each criterium is 1

Partial relevance numbers (relevance of goal j for criterium x)
Local relevance numbers (relevance of goal j at level i)

Condition – Sum of local relevance numbers at one level has to be 1

Cumulative direct relevance number – Relevance of goal j for main goal, whole relevance tree