As with everything - 'crap in = crap out' and yes the devil is certainly in the detail. However, there is a huge difference between "arbitrary parameters" and sufficient approximation for the input to a numerical simulation. A sufficient approximation is intrinsic to Finite Element Analysis since infinite precision and accuracy is not possible.
I must confess that I haven't read the appendix z report fully, I've only read the simulation setup and skimmed the results. Could you point me to the approximations of input parameters that have been made that are arbitrary?
As I've understood: It's stated that the model of the 9N314 warhead that was used is a 70 kg cylindrical charge of 60% RDX and 40% TNT. The "modified fisher" model was used to calculate the energy released as a blast wave. The aircraft was travelling at 254 m/s and the warhead ~600 m/s. You mentioned reconstruction of the trajectory - is a highly accurate trajectory necessary for simulation of the warhead and its interaction with the airframe as per this report?