Course : ME22023 NEW PRODUCT DEVELOPMENT

UNIT 4 DECISION MAKING AND PRODUCT ARCHITECTURE

Unit 4 Part 1 - DECISION MAKING

Dr C Senthamaraikannan
Associate Professor,
Dept of Mechanical Engineering
Sri Venkateswara College of Engineering
 Unit 4 DECISION MAKING AND PRODUCT ARCHITECTURE

• Decision making –decision theory –utility theory –decision trees –concept evaluation
methods–Pugh concept selection method- weighted decision matrix –analytic hierarchy
process – introduction to embodiment design –product architecture – types of modular
architecture – steps in developing product architecture.
 Definition

• Product concept selection is a decision process,

in which the design team selects one or a few
product concept for further development

5/1/2025 3
 Challenges

• How to choose the best concept of abstraction?

• How to embrace all inputs (likings and concerns)
from the whole team in the decision process?
• How to make use of good attributes of
otherwise weak concept designs?
• How to document the decision process?
5/1/2025 4
 Concept Selection Approaches
• External decision

– By use of an external group of customers, clients, etc.

• Product champion & intuition

– By an influential member of the development team

• Multi-voting

– Asking each member to pick a number of concepts and pick the one with most votes.

• Pros and cons

– The team list the strengths and weakness of each concept.

• Prototype and test

– Build and test prototype for each concept and select based on the test data.

• Decision metrics

– The team rates each concept against selection criteria with varying importance/weights.

5
 Two stages of concept selection
• Concept screening (the Pugh concept selection method)
– To quickly narrow the number of concepts and to
improve the concepts
• Concept scoring
– weighs the relative importance of the selection
criteria
– focus on more refined comparisons with respect to
each criteria

5/1/2025 6
 Concept Selection Process
1. Prepare the Matrix
– Criteria
– Reference Concept
– Weightings
2. Rate Concepts
– Scale (– 0 +) or (1–5)
– Compare to Reference Concept or Values
3. Rank Concepts
– Sum Weighted Scores
4. Combine and Improve
– Remove Bad Features
– Combine Good Qualities
5. Select the Best Concept
– May Be More than One
– Beware of Average Concepts
6. Reflect on the Process
– Continuous Improvement

5/1/2025 7
Concept Development Funnel

concept generation

concept screening

concept scoring
concept testing
 Remember…
The goal of concept selection is not to
• Select the best concept.

The goal of concept selection is to

• Develop the best concept.

So remember to combine and refine the

concepts to develop better ones!
 Caveats
• Beware of the best "average" product.

• Perform concept selection for each different customer group and compare results.

• Check sensitivity of selection to the importance weightings and ratings.

• May want to use all of detailed requirements in final stages of selection.

• Note features which can be applied to other concepts.

Concept Selection Example:
Reusable Syringe
 Mission Statement
• Product description:
– Reusable syringe with precision dosage control for
outpatient use.
• Primary market
– Elderly
• Major features:
– Accuracy of dose metering
– Inexpensive

5/1/2025 12
 Need analysis
• Ease of handling
• Ease of use
• Readability of dose settings
• Accuracy of dose metering
• Durability
• Ease of manufacture
• Portability
5/1/2025 13
 Concept screening steps
1. Prepare a selection matrix based on
the selection criteria
2. Rate the concepts
3. Rank the concepts
4. Combine and improve concepts
5. Select one or more concepts
6. Reflect on the results and the
process
5/1/2025 16
 Concept screening principles
• Be focused on customer needs
• Match or exceed competitors’ performance along key
dimensions
• Improve the product’s manufacturability
• Reduce lead time
• Encourage more and effective participation from the
design team members
• Have better documentation of the decision process.

5/1/2025 17
Example: Concept Screening
5/1/2025 20
5/1/2025 21
 WEIGHTED DECISION MATRIX
▪ A weighted decision matrix is a method of evaluating competing
concepts by ranking the design criteria with weighting factors and
scoring the degree to which each design concept meets the criteria.
▪ To do this it is necessary to convert the values obtained for different
design criteria into a consistent set of values.
▪ The simplest method of dealing with design criteria expressed in a
variety of ways is to use a point scale.
▪ A 5-point scale is used when the knowledge about the criteria is not
very detailed. An 11-point scale (0–10) is used when the information is
more complete (Table ). It is best if several knowledgeable people
participate in this evaluation.
 Concept scoring steps
1. Prepare a selection matrix, based on
the selection criteria
2. Rate the concepts
3. Rank the concepts
4. Combine and improve concepts
5. Select one or more concepts
6. Reflect on the results and the
process
5/1/2025 23
 Concept Scoring Method

n
S j =  rij wi
i =1

– where
• wi = the weight for the ith criterion
• rij = raw rating of concept j for the ith criterion

5/1/2025 25
 Scale for Rating the Concepts
(relative performance)
1: much worse than the reference
2: worse than the reference
3: same as the reference
4: better than the reference
5: much better than the reference

5/1/2025 26
Example: Concept Scoring
 Issues in the decision process
• It assumes that the selection criteria reflect customer’s needs
• It assumes that the selection criteria are independent
• Manufacturing costs and manufacturability are not included in the
decision
• It is better to directly evaluate those independent, simpler concepts
underpinning the design concepts, if they constitute all the product
concepts.
• The process of concept selection can be applied to throughout the
development process.

5/1/2025 28
Concept Selection Exercise:
Mechanical Pencils
 What are the screening and scoring criteria for
the following pencil holder concepts?

5/1/2025 31
 ANALYTIC HIERARCHY PROCESS
▪ The Analytic Hierarchy Process is a problem-solving methodology for making a

choice from among a set of alternatives when the selection criteria represent

multiple objectives.

▪ AHP was developed by Saaty.1 AHP builds upon the mathematical properties of

matrices for making consistent pairwise comparisons.

▪ Not only is AHP mathematically sound, but it is also intuitively correct

ANALYTIC HIERARCHY PROCESS
 ANALYTIC HIERARCHY PROCESS
Calculating Geometric Means:
•The next section calculates the geometric mean for each row of the normalized matrix.

•The geometric mean is a type of average that is useful when dealing with ratios or
multiplicative relationships. For a set of n numbers, the geometric mean is the n-th root of
the product of those numbers.
•For Price (P):
•wP​=(0.65×0.69×0.56)1/3=(0.25046) 1/3≈0.63
•For Mileage (M):
•wM​=(0.22×0.23×0.33) 1/3=(0.016722) 1/3≈0.26
•For Ground clearance (G):
•wG​=(0.13×0.08×0.11) 1/3=(0.001144) 1/3≈0.05
 ANALYTIC HIERARCHY PROCESS
Normalizing the Weights:
•The final step is to normalize these geometric mean weights so that they add up to 1 (or
approximately 1 due to rounding).
➢ Sum of the geometric means: 0.63 + 0.26 + 0.05 = 0.94
➢ Normalized weight for Price: 0.63 / 0.94 ≈ 0.67
➢ Normalized weight for Mileage: 0.26 / 0.94 ≈ 0.28
➢ Normalized weight for Ground clearance: 0.05 / 0.94 ≈ 0.05
Interpretation:
The final normalized weights (0.67 for Price, 0.28 for Mileage, and 0.05 for Ground
clearance) represent the relative importance of each car attribute based on the initial
pairwise comparisons.
In this example, Price is considered the most important factor (67%), followed by Mileage
(28%), and then Ground clearance is considered the least important (5%).
This method, often part of the Analytic Hierarchy Process (AHP), helps in making decisions
by structuring complex problems into a hierarchy and comparing elements pairwise.
Dr C Senthamaraikannan
Associate Professor,
Dept of Mechanical Engineering
Sri Venkateswara College of Engineering