The maximum distance between the object and the tube will be 30cm in my case, as irradiance is inversely proportional to distance, I am simply multiplying the irradiance value by 3, so I get 135 uW/cm2.
I would expect irradiance being inversely proportional to the square of the distance, so you should multiply by 9, to get 405 uW/cm
2.
When you later multiply by 6, that could be big underestimate, since if your face of the object is 30cm away from one of the lights, it probably is quite closer to another, and if the closer one is at say 15cm, then it counts as 4 sources at 30cm, due to the inverse square law. Multiplying by 6 could also be a overestimate, because some lightsources (tipically half?) should be occluded (or not facing directly), so only half reach your face of the object, and probably not at a right angle. Then again, the walls are reflective.
I think a good idea is to consider first how will you place the light sources, which are not point-like, and then to get the lower bound of irradiance, try to imagine the face of a
typical object that will be farthest (in the inverse square sense) from all light sources, and the more ocluded / facing away. That will be your lower bound. And that doesn't consider the reflectivity of the walls.
Maybe the best advice for your project I can give: try using 4-pin CFLs, because there are goood, compact and efficient electronic ballasts for them, which also give a longer life.
I've put together something like your project recently, but much more crude, with just one 4W Philips CFL. Those two pin CFLs have an internal ballast, so you just need to limit the mains current going into the lightbulb with some kind of inductance. Since I didn't own a limiter, I improvised one with two 12V transformers back to back. They get hot, but get the work done. It would have been much better if I had bought four pin CFLs with the correct electronic ballasts. I was under very strict lockdown then, so I had to stick to this scheme.
Edit: another thing about my crude project. I wanted it to sterilize masks. Just leaving the mask lying in the box doesn't work, because there are always parts of it in complete shadow. So I considered (but didn't have time to) using a small synchronous motor from a microwave oven to make a rotating stand for the mask. That way I wouldn't need to manually turn it several times to make sure it is disinfected.