Final Flashcards
Doe
An approach for effectively and efficiently exploring the cause and effect relationship between numerous process variables and the output or process performance variable. Identifies the vital few sources of variation
Root cause approaches
1 observe the process
2. Experiment with the process
One factor at a time approach
Change one factor at a time.it is impossible to tell if the variables interact with each other
Factorial approach
Changes several factors at a time, initially begins with only 2 level for each factor. Handles common cause variation easily
Why randomize runs?
To eliminate the effect of lurking variables
Why replicate the runs?
To measure the amount of inherent variation or pure error. Replication
Replication
Sample size , repeating all the experimental conditions two or more times
Randomization
To assign the order in which the experimental trials will be ran using a random mechanism to average the effect of any lurking variable
Lurking variable
One that has an important effect and yet is not included among the factors under because its existence is unknown, thought to be negligible, data on it are unavailable
Two types of effects
- Main factor effect: overall effect of each factor on the response
- interaction effect: synergy between the positive or negative factors
Interaction effect
Effect of one factor has on the response is not the same
Two level experiments
2 to the k, where k is the number of factors
Center points
Detect non-linearities, estimate quadratic effects for quantitative factors
X and MR chart
N= 1, variable.x measures location and Mr is moving range or dispersion
X,R chart
N is less than 10