ok, primary load Fz ... apply to the two web fasteners, react on the other side of the split. Remember there is now a couple ... if you like the Fz load at the split is being carried by the two web rivets "d" inches away from the split, so there's an Fx couple ... Fx*h (distance between web rivets) = Fz*d ... yes? clear as mud ?
next primary load Fx ... apply over all 6 rivets, or 4 cap rivets, or 2 web rivets !!??
Next primary load, Mz ... this is a couple between the inner and outer finger splices. The splice is H inches high (between the finger straps), then load into caps is Mz/H over 2 rivets.
Fy, Mx and My should be small, and out-of-plane ... neglect or apply to skin rivets.

But better is to look up how your company wants you to analyze splices (rather than listening to some guy on the internet).

another day in paradise, or is paradise one day closer ?

That's a poor way to splice a longeron. The splices cannot transfer shear loads between each other. I definitely would never do that, and never have. Use angles or something other than just plates.

Some feedback on your assumption: "... as shear forces on the web and tension forces on the caps."
If you were to compare the relative stiffness of those two loadpaths (in-plane shear vs out-of-plane tension) you'd find those joints to be an order of magnitude (or more) softer out-of-plane, compared to in-plane. Hence why fastener tensile loading is often considered neglible when idealising structures.
In summary, think about relative stiffneses. Loads follow the stiffest loadpath. So you shouldn't expect any significant load to be carried by fastener tension when there's an adequate shear connection available.

Hi all

Thanks for all answers. As far as i understand from your answer.

I have to solve the problem with in plane loads.All the fasteners should carry the shear loads. Am i right?