I'm not quite sure where you're going ...
assuming you have multiple stiffeners dividing the tank shell. the stiffeners allow the shell to have discrete changes in shear stress (shear flow) ... as opposed to a shell which would have a continuously varying shear stress. and these different shear flows require an endload in the stiffeners to react them.
it'd make sense to me that you could start with the unstiffened shell to get a nominal shear stress distribution, and then discretise this over your several bays. i think you'll find a difference proportional to the distance from the neutral axis (so that the stiffener endload resembles a bendins stress field).
you're probably working on shear stress = VQ/It, where Q is the 1st moment of area of the cross section above (away from the section neautral axis) the section plane (where you're calculating the shear stress).
hopefully this isn't as fibberish as it sounds (as i read it !)