One primary reason software isn't a true solution, is that it can't tell the difference between plants and non-plants. Software will, for example, apply the IR effect to a green car or building as well as green plants because it doesn't have the complete picture when it comes to spectral components of green paint versus a living green plant. (giving myself the thumbs up for that pun )
IR photography can be used to study variations in vegetation because it not only sees living green and inanimate green differently, it also sees a difference between things like grass, trees, crops, soil, moisture content, etc. As an example, this enables farmers and the US Department of Agriculture to monitor crop health, moisture and compliance based on their respective spectra.
I think software can make nice IR-like images, but I would think it very difficult to make truly accurate conversions.
In addition to the effect of sensor filters, IR wavelengths are different enough from visible light to affect focus. One often gets soft images when using an on-the-lens IR filter; so you may need to make adjustments for that. Many lenses have IR adjustment marks on the focus ring. Even if you do a camera conversion, this is still an issue because the lenses and focus system are designed for visible light.