What have I done wrong with shuffle optimization, why the core ply change the position, actually it should not be stacked?

I am doing a shuffle optimization with monocoque structure with composite material, and the result of it did not correct. The core ply (Rohacell) should stay still. I do not know which steps did I miss? 

As you can see the core ply which has orientation on 60 degree need to stay still as iteration 0 but the other plies should stack until the optimizer find and compute the optimal layout for this structure.

The result that I need to get will be like following picture.

I will attach the model file of my shuffle optimization, so you can check it out.

Thank you in advance. Please help me I am struggled with this for 3 days.