Concept Generation, Concept Selection, & Concept Testing

ENME 371 Product Engineering & Manufacturing

ENME 371 - C. Thamire

3/14/2012

-0-

Announcements and Outline

Announcements
Quiz 3 Mid-term Exam on 3/28

Lecture Topics
Concept Generation, Selection, and Testing

-1ENME 371 - C. Thamire

Next Steps in the PDP Process

Problem Formulation Product Specification Concept Generation Concept Selection System Design Detail Design Fabrication, Testing, and Modification Design and Build Tooling Production and Distribution Tracking and Assessment
Service Center

-2-

3/14/2012

Concept Generation and Selection

Functional Decomposition Concept Generation Concept Selection Concept Testing
Constraints These are limits in design Ignore constraints completely or relax them at first Leads to a very large set of possible solutions initially Reimpose constraints later remove infeasible concepts (too heavy? Too weak? ...)
-3-

Criteria These are design objectives (minimize weight, improve performance, ) Apply and choose the optimum concept

3/14/2012

Concept Generation
Define subsystems Find existing Concepts Generate new concepts Eliminate poor ideas Prepare Design Proposals

Design can be .. Ill defined Recursive Open ended

Some Popular Methods

Generate ideas in a brain-storming session Reasoning by Analogy Get ideas from others
-43/14/2012

Concept Generation
Some Formal Concept Generation Methods
Morphological Approach - divide into sub-problems, generate concepts, combine concepts into complete solutions Synectics analogies Theory of Problem Solving (TIPS/TRIZ) - systematic engineering procedure

ENME 371 - C. Thamire

3/14/2012

From: Pahl et al., Engineering Design: A Systematic -5Approach, 1996 [5].

Some Simple Concept Generation Methods

1. http://www.uspto.gov/ http://www.google.com/patents http://scilib.ucsd.edu/howto/guides/find patents.html http://www.us-patent-search.com/

Using existing ideas Reverse Engineering Patents

2. 3. 4.

5. www.pat2pdf.org Traditional Library: Books (Encyclopedias (Enc. 6. http://www.freepatentsonline.com/ Britannica, Enc. Americana, McG.-Hill Enc. Of Sci. and 7. http://www.delphion.com/ Tech., ) and Subject-related text books (focused 8. http://web.mit.edu/invent/inventsearch)), Journals, Science-Citation Index, main.html

Electronic Library: Internet (Basic-search Engines and On-line Libraries) Reasoning by Analogy

ENME 371 - C. Thamire

3/14/2012

-6-

Example Product
(All materials in this section, including illustrations, are taken from Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

3/14/2012

Example Product: Reusable Syringe

Goal: To develop a reusable syringe with precise dosage control for outpatient use Current Problems: cost (original product was made w/ s.s.) and accuracy of dose metering

ENME 371 - C. Thamire

3/14/2012
(Source: Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

-8-

Criteria Established
Ease of handling Ease of use Readability of dose settings Dose metering accuracy Durability Ease of manufacture Portability

ENME 371 - C. Thamire

3/14/2012

-9-

Concepts

ENME 371 - C. Thamire

3/14/2012
(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

- 10 -

Next Step Concept Selection

3/14/2012

Concept Selection
Concept Screening
- Select the top one-third concepts using the Pugh matrix - Combine and improve the concepts

Concept Scoring
- Weigh criteria - Select the final concept using the concept scoring matrix

ENME 371 - C. Thamire

3/14/2012

- 12 -

The Concept Screening Matrix/ Pugh Matrix

It is recommended that the net scores not be displayed unless the PDP team is confident of near equivalent weights of the criteria.

(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).
ENME 371 - C. Thamire

Note: A B/W/S Scale is also used some times; 3/14/2012 B = Better; W = Worse; S = Same

- 13 -

New and Revised Concepts

(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

ENME 371 - C. Thamire

3/14/2012

- 14 -

New and Revised Concepts

(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

ENME 371 - C. Thamire

3/14/2012

- 15 -

Weighting Factors
The Pugh method assumed all criteria had equal weight But, as you gain more knowledge about your design you may realize that this is not true For example, should serviceability rate as highly as functionality?
ENME 371 - C. Thamire

3/14/2012

- 16 -

Concept Scoring
Step 1: Prepare weights for the criteria Step 2: Prepare the selection matrix Step 3: Rate the concepts Step 4: Rank the concepts Step 5: Select one (or more concepts) Step 6: Reflect and process

ENME 371 - C. Thamire

3/14/2012

- 17 -

Weighting Factors

Good Rule: Make sum of the weighting factors equal to unity.

n

w
i =1

= 1.0

0 wi 1

where n is the total number of evaluation criteria

Three ways to do this

ENME 371 - C. Thamire

3/14/2012

- 18 -

Weighting Factors
First way to do this:
1. Assign 100 points between different criteria 2. Normalize weights by dividing each criterion by 100 based on information from HOQ and any additional information

ENME 371 - C. Thamire

3/14/2012

- 19 -

The Concept Scoring Matrix

(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

ENME 371 - C. Thamire

3/14/2012

- 20 -

Weighting Factors
Second way: Pairwise Comparison (Use for our
project!)
1. Compare each criterion with others. 2. Decide which is important. 3. Assign a 1 to the more important criterion and a 0 to the other. 4. Sum the values assigned to each criterion (row total) to determine its relative importance 5. Normalize by dividing by the sum of the row totals to give the weighting factor.
ENME 371 - C. Thamire

3/14/2012

- 21 -

Weighting Factors

Second way: Pairwise Comparison

Criteria A B C D A 0 1 1 B 1 0 0 C 0 1 1 D 0 1 0 Row totals 1 2 1 2 6 Weight factor 0.17 0.33 0.17 0.33 1.00

ENME 371 - C. Thamire

3/14/2012

- 22 -

Example Concept Selection Problem

ENME 371 - C. Thamire

3/14/2012

- 23 -

Weighting Factors

Third way: Objective tree

Useful in advanced stages of design (when decisions are made based on technical as well as less subjective criteria) For example, you may want to choose among materials or motors, on the basis of performance data, while at the same time including durability or time to make a replacement part.
ENME 371 - C. Thamire

3/14/2012

- 24 -

Weighting Factors

Third way: Objective tree

Example: Choose between three designs for making a large steel crane hook for hoisting heavy ladles of molten steel. The three designs are: 1. Built-up with welded steel plates 2. Built-up with riveted steel plates 3. A monolithic cast steel crane hook
ENME 371 - C. Thamire

3/14/2012

- 25 -

Weighting Factors by Objective Tree

Construct a hierarchical objective tree. Start with the crane hook and successively break it down into finer and finer criteria. Sum of the weights must add to 1.0.
Crane Hook 1.0 Cost(0.6) Quality in Service (0.4) Repair(0.2) 0.6x0.2=0.12 Durability(0.6) 0.4x0.6=0.24
3/14/2012

Mfg. Cost (0.5) 0.6x0.5=0.30 Matl. Cost (0.3) 0.6x0.3=0.18

ENME 371 - C. Thamire

Reliability(0.3) 0.4x0.3=0.12 Prod. Time (0.1) 0.4x0.1=0.04

- 26 -

Crane Hook Example

Criteria: 1. Material cost 2. Manufacturing cost 3. Ease of repair 4. Durability 5. Reliability 6. Time to make a replacement crane hook 3 Concepts generated!
ENME 371 - C. Thamire

3/14/2012

- 27 -

Weighted Decision Matrix

First take the magnitude of the design criterion and apply the best appropriate scale factor. Multiply this score by the weighting factor.
D esign criterion Weight factor Units

Built-up plates w elded

Magnitud e Score 8 7 7 8 7 7 Rating 1.44 2.10 0.84 1.92 0.84 0.28 60 2500 Good H igh Good 40

Built-up plates riveted

Magni -tud e 60 2200 Excel. H igh Excel 25 Score 8 9 9 8 9 9 Rating 1.44 2.70 1.08 1.92 1.08 0.36

Cast steel hook

Magnitud e 50 3000 Fair Good Fair 60 Score 9 4 5 6 5 5 Rating 1.62 1.20 0.60 1.44 0.60 0.20

Material cost Mfg. cost Repair Durability Reliability Tim e to m ake

0.18 0.30 0.12 0.24 0.12 0.04

/lb $ Exper Exper Exper H ours

7.42

8.58

5.66

ENME 371 - C. Thamire

3/14/2012

- 28 -

Evaluation of Design Parameters

Parameters have different values and units. Need an evaluation scale to put things on the same basis Use a 5-point scale (0-4) when knowledge about the parameters is not very detailed, or an 11-point scale when the information is more complete.
11-point Scale 0 1 2 3 4 5 6 7 8 9
ENME 371 - C. Thamire

Description

5-point scale 0 1

Description Inadequate Weak

10

Totally useless solution Very inadequate solution Weak solution Poor solution Tolerable solution Satisfactory solution Good solution with a few drawbacks Good solution Very good solution Excellent (exceeds the requirement) Ideal solution 3/14/2012

2 3 4

Satisfactory Good Excellent

- 29 -

Concept Testing
(From Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995 [2]).

3/14/2012

Concept Testing
When?
Identify customer needs Establish Target Specifications Generate Product Concepts Select Product Concept(s)

Development Plan

Plan Downstream Development

Set Final Specifications

Test Product Concept(s)

Cannot test more than a few concepts with customers so must narrow concepts first!

ENME 371 - C. Thamire

3/14/2012

- 31 -

Concept Testing - Why?

Solicit response to product concept(s) Select the final concept Improve the final concept Estimate the sales potential of the concept

No testing when: testing times are high testing costs are large
ENME 371 - C. Thamire

3/14/2012

- 32 -

Concept Testing - Steps

Define purpose Choose population Choose survey format Communicate concepts Measure responses Interpret results Reflect and process

ENME 371 - C. Thamire

3/14/2012

- 33 -

Example Product
A three-wheeled electric-powered scooter that could be folded and carried easily Need to assess the concept to decide whether to proceed or not

ENME 371 - C. Thamire

3/14/2012

- 34 -

Step1 - Define Purpose

Which of the alternative concepts? How to improve to better meet needs? How many units will be sold? Continue development?

ENME 371 - C. Thamire

3/14/2012

- 35 -

Step2 - Choose Population

Does population reflect target market? Biased population? Multiple target markets (college/urban)? Sample size?

Multiple surveys:
Smaller for early test Larger later
ENME 371 - C. Thamire

3/14/2012

- 36 -

Step 3 - Survey Format

Face-to-face interaction - intercepts, prearranged interviews, trade-show booth, focus groups,.. Telephone - specific individuals Postal mail Electronic mail Internet

Risk of bias!
ENME 371 - C. Thamire

3/14/2012

- 37 -

Step 4 - Concept Communication

Verbal Description - paragraph or summary of points, read by customer or surveyor Sketch - Line Drawing Photos and Renderings Story Board - Series of images conveying temporal sequence Video Simulation Interactive Multimedia Physical Appearance Models Working Prototypes
ENME 371 - C. Thamire

3/14/2012

- 38 -

Step 4 - Concept Communication

Verbal Description

The Product is a light-weight electric scooter that can be easily folded and taken with you inside a building and public transportation. Weighs 25 pounds Travels at speeds of up to 15 miles/hr Can go about 12 miles on a single charge Can be recharged from std. Electric outlets Easy to ride and control - has only an accelerator button and a brake

ENME 371 - C. Thamire

3/14/2012

- 39 -

Step 4 - Concept Communication

Sketch
Line Drawing

ENME 371 - C. Thamire

3/14/2012

- 40 -

Step 4 - Concept Communication

Photos and Renderings

ENME 371 - C. Thamire

3/14/2012

- 41 -

Step 4 - Concept Communication

Story Board - Series of images conveying
temporal sequence

ENME 371 - C. Thamire

3/14/2012

- 42 -

Step 4 - Concept Communication

Physical Appearance Models

ENME 371 - C. Thamire

3/14/2012

- 43 -

Step 4 - Concept Communication

Working Prototypes
Risky -May perform better -May perform worse -Less attractive

ENME 371 - C. Thamire

3/14/2012

- 44 -

Step 5 - Measure responses

Ask to choose from two or more concepts Ask why he/she favors that one Ask how to improve Rate
Definitely would buy Probably would buy Might or Might not buy Probably would not buy Definitely would not buy
ENME 371 - C. Thamire

3/14/2012

- 45 -

Step 6 Interpret Results

If results are conclusive, choose that concept, provided customers understood the product Otherwise, offer a concept based on cost or other considerations Estimate the sales potential

ENME 371 - C. Thamire

Absent past history, take Cdefinitely = 0.4 and Cprobably = 0.2 3/14/2012

- 46 -

Step 7

Reflect and process

ENME 371 - C. Thamire

3/14/2012

- 47 -

Conclusion
In this lecture, we have:

Introduced you to Concept generation and selection Process Next class: Product Architecture

References
1. 2. 3. 4. 5. Schmidt, Zhang, Herrmann, Dieter, and Cunniff, Product Engineering and Manufacturing, College House Enterprises, LLC, Knoxville, TN. E. B. Magrab, Integrated Product and Process Design and Development, CRC Press, New York, 1997. Ulrich and Eppinger, Product Design and Development, McGraw Hill, NY, 1995. J. R. Dixon and C. Poli, Engineering Design and Design for Manufacturing, A structured Approach, Field Stone Publishers, MA, 1995. Pahl, G. and W. Beitz, ., Engineering Design: A Systematic Approach, SpringerVerlag, London,1996.
3/14/2012

ENME 371 - C. Thamire

- 48 -