Implementing the DF4 in a robust model, allowing for enhanced comparison, prioritisation and grouping of...

It is here shown how partial order can be used to provide a robust and consistent implementation of the DF4 approach which provides unbiased information enabling comparison and open up the possibility for grouping. The approach is based on few assumptions, works well with the data, can include different types of input parameters, and can provide fundamental information about the ranks of tested materials. It is shown that the materials in many cases are below one threshold within a tier, but above another threshold within the same tier. It is also observed that the ranks of the materials can differ between tiers, although this is less relevant for DF4 since parameters evaluation may be hierarchical.