1. Analytic Hierarchy Process
(A.H.P.)
P.V.for all criteria
P.V. for
Criterion #1
P.V. for
Criterion #2
P.V. for
Criterion #3
P.V. for
Criterion #4
TOTAL SCORE
Design #1
Design #2
Design #3
2. Outline
- What is Analytic Hierarchy Process (A.H.P.)?
- What is the need that (A.H.P) can cover?
- How does (A.H.P.) fit into the design process?
- Methodology
- Example
- Conclusions / Wrapping up
3. School of Mechanical, Aerospace &
Automotive Engineering
What is Analytic Hierarchy Process (A.H.P.)?
5036MAA – Design and Sustainability
A.H.P.:
is a multi-criteria decision making method
can be used to rank a set of criteria (major design objectives)
can be used to rate the criteria on a relative scale of importance
can be used to rank & rate alternative designs
allow some small inconsistency in judgment because (human
judgement is not always consistent)
Main idea:
to introduce ratio scales for paired comparisons
to form a normalised comparison matrix
to derive a priority vector for each comparison matrix
to use priority vectors to determine relative importance of designs
4. Analytic Hierarchy Process: used
to (a) rank criteria to be included
in our design; and (b)
rank alternative designs
according to their relative
importance
5. AHP: Priority Vector
Main task: to rank alternative designs according to their relative importance
Level 0: goal of the analysis
Level 1: multi-criteria to be used (can also add other levels of sub- criteria)
Level 2: alternative choices (alternative designs)
Example:
Factor A: Cost; Factor B: Quality; Factor C: Performance; Factor D: Manufacturability
Choice X: Design #1; Choice X: Design #2; Choice X: Design #3
6. CRITERI
A
Cost Quality Performance Manufacturability
Cost 1
Quality 1
Performance 1
Manufacturability 1
CRITERI
A
Cost Quality Performance Manufacturability
Cost 1 2 3 3
Quality 1 3 2
Performance 1 2
Manufacturability 1
Step 2: Use own judgement and complete upper-triangular part of matrix
CRITERI
A
Cost Quality Performance Manufacturability
Cost 1 2 3 3
Quality 1/2 1 3 2
Performance 1/3 1/3 1 2
Manufacturability 1/3 1/2 1/2 1
Step 3: Complete lower-triangular part of matrix (reciprocal of upper-triangular part)