This is quick and easy, right? - Read This First and Foremost 

Artificial sky studies are very unlike heliodon studies. They require a much more deliberate, purposeful and designed approach to testing and predicted outcome. Testing requires a specific plan with a specific set of alternatives to be tested. Models intended for artificial sky testing must be constructed specifically for this purpose and must pay great attention to certain details. Model construction and testing must adhere to a stringent set of criteria and rules to produce meaningful results. While this is not intended to make artificial sky testing seem impossible or unworthy of the effort, it is meant to state that meaningful artificial sky testing is NOT a casual exercise to be entered into without understanding and preparation. Unlike heliodon testing, it is not possible to bring down any existing design study model and "play and experiment" to learn about the overcast sky daylighting performance of a proposed design.

Why am I doing this again? - Testing Plan: What Will Be Learned

Because of the preparation required, the time investment in testing and the specific procedures involved, it is best to have a very clear idea of the question(s) to be answered with a testing exercise. The very nature of the question can guide the construction of the model and determine the appropriate testing variables and procedures. Careful planning and execution is required for meaningful artificial sky testing.