Run the Design

Run the Experiment (Trial Batches)

The next step after generating an experimental design is to actually perform the experiment. The experiment in this case is a set of trial batches from which specimens will be fabricated and tested for the responses and mixture components specified in Steps 1 and 2. At this point you should have completed Steps 1 and 2 and obtained a printed copy of the experimental plan (mix proportions for the set of trial batches). If you have not done so, click on the appropriate selections on the left hand side bar.

If you have completed Steps 1 and 2 but have not printed a copy of the experimental plan, click here to view or print a copy of the experimental plan.

Step 3 is the most time-consuming of the 6-step procedure in that you must physically run the experimental plan and collect data on the performance variables of interest to you. Running the experiment includes the following tasks:

ordering and preparing the materials needed for the experiment

ensuring that equipment is available and ready for use

coordinating personnel

batching, fabricating, testing and recording data (running the experiment)

IMPORTANT! If you are ready to run the experiment, please read the following sections which describe important considerations to keep in mind while performing the experiment

Nuisance Factors and Run Sequence Randomization
Running the Experiment

1. Nuisance Factors and Run Sequence Randomization

Your experiment has three types of factors:

Major factors (variable factors)
Fixed factors
Nuisance factors

The primary goal of your experiment is to determine optimal settings of the major factors that were selected as variable factors in Step 1. You may have also designated some fixed factors to be held constant throughout the experiment.

The major factors and fixed factors are controlled in the experiment. However, in addition to these, there are other factors which are not controlled in the experiment but which could possibly effect the experimental results. These are called nuisance factors. It is often assumed that these nuisance factors should not have any effect, but in reality they may have an effect.

Nuisance factors may include the following:

mid-experiment changes in
- instruments
- equipment
- environmental conditions (temperature, pressure, humidity, etc.)
- measuring devices
- test procedures/protocols
day of week
time of day
operators

Nuisance factors may affect the measured test results, which would in turn affect the data analysis, and ultimately the final conclusions (i.e., the estimated values for the optimal mixture proportions).

Run sequence randomization is used to minimize the effect of nuisance factors. Experiment designs are usually generated in a "standard order" based on the settings of the factors. This order, which is used in the data analysis, is indicated by the "mix number" column (column 2) in the experimental plan generated by COST. Run sequence randomization is the general experiment design technique in which random numbers are assigned to each of the specified runs in the experiment, and these random numbers determine the order in which the experiment is to be run (the "run order" or "run sequence"). The experimental plan generated by COST is printed in run order -- the first column of the experimental plan, labeled "run number", is the run order to be used for the experiment.

It is very important to follow this run sequence in order to minimize possible error in the experimental results caused by known or unknown nuisance factors in the experiment.

2. Running the Experiment

The quality and accuracy of the final mixture proportion settings will depend very much on the care taken in carrying out the experiment. The following is a list of recommended practices:

Attention to detail and proper execution in batching, mixing, fabricating, curing, testing, and recording results are essential.
For each trial batch, use several (preferably 3 or more) specimens for each test.
The experimental plan contains 3 or 5 center point runs. These are replicates (repeats of batches using the same settings) which can be used during the experiment as control mixes. Scheduling to allow for one control mix per week is recommended -- in this way significant week-to-week variation can be detected.
If possible, the same operator should perform the same tasks throughout the experiment.
Mistakes will inevitably occur even in the best laboratories, and it is important to acknowledge this and to be prepared to repeat a batch if it is suspected that an error has occurred.

Return to top of page