Shell Stress output very high

[marvelous7]

Hi All,

I'm modeling a 200 m long (10.5 m wide) pavement in SAP2000 using Membrane area elements (1 m x 2.1 m). The area elements are assigned area springs to simulate friction. The material is loaded in tension. The material model used is an area average of steel and concrete models in tension as an approximation. It is provided using the user-input option. The max. strength in tension is around 1000 psi. However, after analysis, the output stress plot shows all points at a higher stress level. I cannot understand how is that possible.
The loading consists of:
End displacement of -1.38" at the left end
Temperature decrease of 50.8F

I've also tried using steel (Grade 60) as the material, in which case output stresses are of the order of 10^9 psi. Somehow, the output stresses appear to be much higher than the material strength. I've checked the units in each case. Please suggest how can I fix this.
I've attached my model here.
Thanks in advance.

 

[engineering_patrol]

Don't consider the peak stress values at corners, as they are not realistic.

 

[marvelous7]

What would be the correct way to obtain the stress variation then?
I'm looking at the S11 stress distribution at the final step with no stress averaging. The stresses seem to vary only along the length of the pavement. I didn't observe significantly higher corner stresses.
I tried to plot the output stress-strain for the material, these points seem to lie on the elastic branch of the curve when extended much further. It appears as if the material continues to behave in an elastic manner beyond the input curve elastic region.