iSixSigma

Experiment Setup Help

Six Sigma – iSixSigma Forums General Forums Tools & Templates Experiment Setup Help

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
    Posts
  • #54108

    Toby
    Guest

    I need to setup an experiment to compare textile color loss produced by an accelerated weathering machine and textile color loss due to actual weather. The objective is to find out how many hours in the accelerated weathering machine is equivalent to hours in the field, and possibly what factor has the most impact on color loss. My initial thought was to setup specimens (12 for each fabric type with varying colors) in the southeast, northeast, midwest, and western parts of the US (each facing the same direction and at the same angle). Each month 1 each would be evaluated for color loss and graded according to gray scale. The temperature, humidity, precipitation, # of daylight hours, and UV index would be recorded as a mean for each month.

    First, does this sound right or am I missing something?
    Second, what statistical test would I use to determine a general ration between machine vs. actual? and which factor had the largest impact?

    Any help would be greatly appreciated!

    Thanks!
    Toby

    0
    #193669

    MBBinWI
    Participant

    @toby_macdonald – what you are looking to do is a designed experiment. You have overtly mentioned five factors, and have identified that there is at least one other factor (hidden) in the regional location (sun angle, ozone level, or some other factor). Then that would be correlated with a second data set. There are several problems with how you are looking at approaching this problem. First, you want to use the means by month, but not all months are the same (28 days vs. 31 days) and you will not get the same weather year to year. You would be better off doing several smaller controlled experiments to evaluate which of these factors are sensitive, and which aren’t and whether there are any interactions. Once you have reduced the variables, then look at doing the correlation.

    0
    #193670

    Toby
    Guest

    Thanks for the insight @MBBinWI. I’ll be scaling this down as you suggest and I will be running the experiment every year to capture annual changes. However, once I have some data how do you suggest I relate time in an accelerated weathering machine and time in the field?

    0
    #193673

    MBBinWI
    Participant

    @toby_macdonald – by measured reading.

    0
Viewing 4 posts - 1 through 4 (of 4 total)

You must be logged in to reply to this topic.