In my opinion, the problem is quite complex because it is difficult to model a cardboard tray. It is a non-homogeneous material that has different mechanical characteristics depending on the direction of stress.
Putting the effort corresponding to your 80g pieces is not really a problem:
Solution 1 you put the corresponding force on the surface of each cell (load distributed over the entire surface) it's super simple to do but optimistic unless your parts really cover the entire surface of the cell.
Solution 2: You create 'separation lines' in each cell to apply your force to the surface occupied by the part
Solution 3: a nice assembly with your tray + the parts (and gravity management for everyone). Given the size of your model this should be calculated quite quickly
On the other hand, the important point is: how is your tray maintained. Follow the boundary conditions that you will apply, your results will be totally different.
Note: Your model has symmetry axes. So you can calculate 1/4 of the model if your chainring holds are also symmetrical (or 1/2 if they are not totally symmetrical)
Don't forget the gravity on the board (surely not negligible since your pieces are light and your board is quite large)