EE382 Spring ‘10

By the end of this chapter, you should:

� Have an understanding of the types of projects electrical and computer engineers undertake.

� Understand and be able to apply sound criteria for project selection.project selection.

� Know how to determine, document, and rank end-user needs.

� Be aware of resources for conducting research surveys.

� Have selected a project concept and developed a problem statement.

2

EE382 Spring ‘10

Creative DesignsCreative DesignsCreative DesignsCreative Designs

Represent new and innovative products. An example of a creative design is the Palm Pilot™ PDA.

Variant DesignsVariant DesignsVariant DesignsVariant Designs

Are variations of existing designs, where the intent is to improve performance or add features to an existing system.

3

EE382 Spring ‘10

Routine DesignsRoutine DesignsRoutine DesignsRoutine Designs

Represent the design of devices for which theory and practice are well developed. Examples are DC power supplies, analog and digital filters.

Experimental DesignsExperimental DesignsExperimental DesignsExperimental Designs

The objective is to design experimental procedures and apparatus for determining the characteristic of a system. An example is to design experiments to test feasibility of gigabit Ethernet in military environment.

4

EE382 Spring ‘10

Some of the questions to answer when starting a new project:

� What are you trying to do? Articulate your goals, using absolutely no jargon.

� How is it done today, and limitations of current

5

� How is it done today, and limitations of current approach.

� What is new in your approach, and why do you think it will be successful?

� Who cares? If you are successful, what difference will it make?

EE382 Spring ‘10

� What are the risks and payoffs.

� How much will it cost? How long will it take?

� What are the midterm and final exams to check for success?

6

EE382 Spring ‘10

The example demonstrates how to apply a project selection model using a method known as the Analytical Hierarchy Process (AHP).

Assume you are in a junior design team that has the Assume you are in a junior design team that has the opportunity to select their project from competing project ideas.

7

EE382 Spring ‘10

Step 1: Determine the selection criteria

A – Match to team skills, B – Technical complexity,

C – Creativity, D – Market potential, E – Industry sponsorship.

8

Step 2: Determine the criteria weightings

The team systematically compares each criterion to all others, using the following scale of importance:

1 = equal, 3 = moderate, 5=strong, 7=very strong, 9=extreme.

EE382 Spring ‘10

Criteria A B C D E GM Weight

A 1.00 5.00 5.00 3.00 3.00 2.95 0.47

B 0.20 1.00 3.00 0.33 0.33 0.58 0.09

C 0.20 0.33 1.00 1.00 3.00 0.72 0.12

9

D 0.33 3.00 1.00 1.00 5.00 1.38 0.22

E 0.33 3.00 0.33 0.20 1.00 0.58 0.09

Sum 6.22

A weight of 5 is used if one criterion is deemed strongly more important than another.

A weight of 1/5 is used if one criterion is deemed strongly less important than another.

EE382 Spring ‘10

Step 3: Identify and rate alternatives relative to the criteria

Assume that team identifies 3 potential project ideas: 1 – IEEE-sponsored robot competition, 2 –Industry-sponsored project to design a new test protocol, and 3 – design of an item-finder device

10

protocol, and 3 – design of an item-finder device to help people locate lost items.

Step 4: Compute scores for the alternatives

The decision matrix is shown below

EE382 Spring ‘10

Criteria Weights Alternatives

Project 1 Project 2 Project 3

A (Match skills) 0.47 0.40 0.20 0.40

B (Complexity) 0.09 0.40 0.30 0.30

11

C (Creativity) 0.12 0.45 0.20 0.35

D (Market potential) 0.22 0.05 0.35 0.60

E (Sponsorship) 0.09 0.00 1.00 0.00

Score 0.29 0.31 0.39

EE382 Spring ‘10

Step 5: Review the decision

Project 3 (item finder) is rated the highest among the three choices on the bases of the weights determined by the team members, and it is a good match to the team skills.

12

EE382 Spring ‘10

Customers often come with the problems and solution all wrapped up together.

Ask clarifying questions, analyze, pick apart the request, and focus on the problem, not the request, and focus on the problem, not the solution.

Researcher and practitioners have examined the problem of eliciting needs and have developed a process for obtaining the voice of the customer following five steps.

13

EE382 Spring ‘10

Step 1: Gather raw data from users

This is accomplished via interviews with supervisors, key users, or people from the client organization.

Step 2: Interpret the raw data in terms of needs

14

Step 2: Interpret the raw data in terms of needs

The needs are expressed in terms of what the system must do (a requirement) as opposed to how it is done.

EE382 Spring ‘10

Step 3: Organize needs into a hierarchy

The marketing requirements are organized into a hierarchy of needs arranged from the most general to the most specific in successive levels of detail. Referred to as an objective treeobjective treeobjective treeobjective tree.

15

Step 4: Determine the relative importance of needs

Pairwise comparison (as shown in the previous example) is a good technique for determining relative importance and weighting of needs.

EE382 Spring ‘10

High-Quality

Audio

(0.24)

Portable

(0.62)

Portable Audio

Device

Easy to Use

(0.14)

Low Distortion

(0.67)

Lightweight

(0.33)

Multiple Data

Limited

Controls

(0..25)

16

Portable audio device to

be used by runners

EE382 Spring ‘10

High-Quality Portable Easy-to-Use Weight

Audio

High-Quality Audio 1.00 0.33 2.00 0.24

Portable 3.00 1.00 4.00 0.63

17

Easy-to-Use 0.50 0.25 1.00 0.14

EE382 Spring ‘10

Step 5: Review the outcomes and the process

There is certain amount of subjectivity in this technique for need identification. The objective of this step is to challenge assumptions, fully identify the problem, and make informed decisions.

18

EE382 Spring ‘10

It is important to conduct a thorough research survey while defining the project concept.

� YouYouYouYou need to become the experts on the problem and state-of-the-art in this area.

� If you don’t, then you re-invent the wheel.

19

� If you don’t, then you re-invent the wheel.

EE382 Spring ‘10

� What is the basic theory behind the concept?

� How is it currently being done?

� What are the limitations of the current designs or technology?

� What are the similarities/differences between your

20

� What are the similarities/differences between your concept and current systems?

� Are there existing or patented systems that are relevant to the design?

EE382 Spring ‘10

◦ Internet resources� Many excellent search engines for locating

resources.

◦ Electrical and computer engineering resources� Circuit Cellar, www.CircuitCellar.com� Datasheet Catalog, www.DatasheetCatalog.com

21

� Datasheet Catalog, www.DatasheetCatalog.com� Dr. Dobbs, www.ddj.com� EE Times, www.EETimes,com� Electronic Design Magazine, www.EDNmag.com� Digi-key Electronics, www.digikey.com� Jameco Electronics, www.jameco.com� Sparkfun Electronics, www.sparkfun.com

EE382 Spring ‘10

◦ Government resources� U.S. Bureau of Labor Statistics, http://stats.bls.gov� U.S. Government Official WebPortal: www.usa.gov� U.S. Patent Office, www.uspto.gov

◦ Journal and conference papers� Association for Computing Machinery, www.acm.org

22

� Association for Computing Machinery, www.acm.org� IEEE Xplore Electronic Library, www.ieee.org

EE382 Spring ‘10

� Two parts of the problem statement are the needsand objectives statements.

� The needs statement identify the need for the project and should:� Briefly and clearly describe the need.� Not provide a solution to the problem.

23

� Not provide a solution to the problem.� Provide supporting information collected.� Describe current limitations.

EE382 Spring ‘10

� The objectives statement ranges from one or two sentences to one or two paragraphs and should:� Summarize what is being proposed to meet the

need.� Provide some preliminary design objectives (detailed

requirements are developed later).

24

requirements are developed later).� Provide a preliminary description of the technical

solution, avoiding a detailed description of the implementation.

EE382 Spring ‘10

Portable Aerial Surveillance Needs and Objectives. Abstracted from the PASS Design Report by Andre, Kolb, and Thaler.

Need: Emergencies happen all across the world, all of the time. There are nearly 2,000,000 reported fires in the U.S. every year, and over 90 tactical activations of

25

U.S. every year, and over 90 tactical activations of Pennsylvania’s Special Emergency Response Team. In large cities, emergency organizations often employ helicopters for aerial surveillance. Helicopters either take too long to reach the scene from nearby city or are too expensive to afford. New helicopters cost well over a million dollars with operating costs of $400-$1000 per hour. There is a need for a low-cost aerial device that can provide emergency workers with overhead surveillance on emergency situations.

EE382 Spring ‘10

Objective: The objective of this project is to design a device that will provide emergency workers with a live aerial view of a situation at a cost that small municipalities can afford. The device will deploy rapidly and record and log video. The camera will also include pan and zoom functionality to make identification of victims and suspect easier.

26

functionality to make identification of victims and suspect easier.