If you look at the concordances you will realise that there is a LOT of input data. This is probably what i've spent the most time on, yet I feel like I could spend a lot more time on it.
The way I manage the input data is by using a separate system called the Transport Data System. The TDS has a few major tasks:
You can find the code here, although I wouldn't recommend seriously looking at using it, maybe just extracting the bits that you need.
You can also access the input data that is used in the model here