A scientist would recommend to build a relevant numerical model with well defined parameters of constitutive model of the backfill. I'd rather suggest the observational method - just start the monitoring (in course of the works) and try to establish model parameters on the basis of back analysis. This method is widely used in geotechnics, because the parameters derived from laboratory testing of soil samples or in situ testing (like CPTU) are not necessarily directly suitable in numerical modelling of the soil body.
Hi Jaroslaw Rybak , thank you very much for your reply. Can you please explain what exactly do you mean by 'model parametres on the basis of back analysis'?