Experimental Design

 Purpose of experimentation:
 - create higher quality products at lower cost.

What is an experiment?

Change in Input  → Process →  change in output

Process → machines, materials, methods, people, environment, measurements

What are the important input factors, and how do they influence the output?

1) Improve performance characteristics:

          Identify critical factors

          Reduce scrap and rework

2) Reduce costs

3) Shorten production development time

 

What is experimental design?

Using efficient and effective experimentation procedures to get the most information you can the fastest you can.

 

Most efficient way to collect data?

Use historical or happenstance data

AKA, just use what is already out there.

 

Risks – undocumented/unknown factors present in non-experimental data.

Example:  
Does changing temp in wire bonding process increase the number of defects in the product?

Undocumented factor: 
Lab operator was changing the standard temperature settings without recording altered temperatures.

Change one factor at a time.

example:

Hold Pressure constant, just change temperature. 

Hold temperature constant, just change pressure.

Do the above tables contain all possible combinations of (T,P)? 

No (P2,T1) is missing.

 

What about interactions between factors?

Example: Testing the effects of heart medication and grapefruit juice on health.  (Cannot test grapefruit juice independently of medication, as they interact with one another) 

 

If factors interact with one another,

one-at-a-time testing is inadequate to determine results.

Combination of (P2,T1) is not contained in the one-at-a-time tables.    What if (P2,T1) provides the optimal result?

Full Factorial Design

Test all possible combinations of input factors

-         Works well if there are only a few input factors
-         Impossible if there are many input factors

Example – what if there are 12 different input factors?

12 input factors, each tested at two different values

(T1, T2, P1, P2, % C1, %C2, etc.)

2^12 combinations = 4096 combinations to test! impossible!

Robust Design:

• Eliminate testing factors that cannot be changed or controlled.
• Use statistical tools to determine the most important combinations to test.
  
  

Planning your experiment:

State specific goals, what information do you need in our upcoming friction lab?

Determine:

• influence of surface area on slip angle
• influence of mass on slip angle
• investigate interaction of different tire treads with different road surfaces

Meet with

• Management,
• engineers,
• subject matter experts,
• operators, 
•  analysts

List inputs

• mass  
• surface area
• tire type (bald tire, all-weather tire, bike tire, etc.)
• road surface type (dirt, cement, ice, water, sand etc.) 
• orientation of tire tread with road surface

List of outputs

• coefficient of friction
• Speed of slip (slow creep? fast slide?)

Objective:

• Minimize slip between tire and all road surfaces

How will inputs be measured?

How many levels per input?  (Use mass 1, mass 2, mass 3 etc.?)

How important is each factor?

 *1.18 Tree diagram

*1.4 Design Matrix Table

Conducting the experiment:

•  Adhere to clearly defined and agreed upon operating procedures
• Prepare data sheets in advance ← Do this before our next lab!
• Repetition & error tolerances
• Analyze data, Draw conclusions, make predictions, do confirmatory tests

Causation vs. Correlation

 

1. A causes B
2. B causes A
3. A and B are both consequences of the same cause
4. Some combination of 1, 2, and 3
5. A and B unrelated, correlation is pure coincidence.

http://www.fromquarkstoquasars.com/correlation-vs-causation/

 

First Lab:
  Test grip of different tire types on different road surfaces

Tires:
Road bike tire
Dirt bike tire
Bald car tire
All-weather car tire

Road surfaces:
Wood plank
plastic mat (slippery like ice)
welcome mat (rough like dirt road)

Mass:
Does the weight of the car matter?
add different weights to see effect on slip angle

Surface area:
Contact area between tire and road surface - does this change the slip angle?

Tread orientation:
skidding sideways vs. stopping straight on - does the orientation of the tire tread with the road surface make a difference?

**There are too many combinations and possible tests to complete all of them in one class period.  You will have to determine what questions are most important to you, then decide what data you need, and what combinations you will test.

** Error - you will need to test each combinations at least 3 times to test repeatability, and determine the uncertainty in your measurements.  We will go through standard deviation calculations, mean, mode, average, etc.

HW:  Consider how much time you have, then decide what factors you want to test, and create a data sheet in excel. 

Alternative experiment if you do not want to test tires - test different types of shoes! (Dress shoes, tennis shoes, sandals, slippers, etc.)