Dear Ata,

I am not an expert in polymers but, in my experience, true stress and strain differ from engineering stress and strain when fracture does not occur suddenly, just at the end of the elongation test, but is the final stage of a damage process that characterizes the entire test. You can use true or engineering curves depending on how you model your polymer: if you are able to take into account how the progressive damage changes the shape of the polymer, you must use the true stress–true strain curve, while, if not, you must use the engineering stress–engineering strain curve.

I made something similar with concrete tested in uniaxial compression: I obtained the true stress–true curve here (I called it effective stress–effective strain relationship):

https://www.researchgate.net/publication/226140897_Experimental_procedure_for_verifying_strain-softening_in_concrete

then I modelled the compression test by using my true curve and uploading the shape of the specimen at each step of crack propagation:

https://www.researchgate.net/publication/235998938_Shape-Effect_in_the_Effective_Laws_of_Plain_and_Rubberized_Concrete

For details on the numerical technique, you can read this further parer:

https://www.researchgate.net/publication/235998931_Crack_propagation_modeling_by_remeshing_using_the_cell_method_CM

Best regards,

Elena