When I did forming simulation, I found that a material failure was predicted earlier then expected, what might be the problem?

One of the common mistakes resulting in early necking is that the bad stress-strain curve is used. Usually, the stress-strain curve is obtained from uniaxial tension test, however, for this test, the obtained stress after diffuse necking might not be dependable. [Instead a stress strain curve can be generated from a biaxial bulge test which enable curves to be obtained out of 70% strain – Mark Finn at Alcan]

How to extrapolate the stress is very important to sheet metal forming analysis and spring back prediction.