As you finish up your last experiments of your PhD research, or put the last pieces of the puzzle together of a theoretical model, you will start thinking: "Now I should start writing..." - a question that probably will echo on and on in your head for the next couple of weeks (or maybe even months) as "But really, once I'm done with XXX, I really will start writing...".
And then one day, you decide to take the plunge and Start Writing. You open a fresh Word document, maybe put some reference papers on your desk and then... then what?
You might stare at your empty screen. You might chew on your pencil. You might get sidetracked and do some busy work first. But the all-encompassing thought in your mind now becomes: "So I decided to start writing - without knowing how to actually Start Writing!".
When you need to produce a book-style dissertation of about 100,000 words, you don't just sit yourself down in front of your computer, write the first line of your introduction, and then look up at the moment that you finish the last sentence of your conclusions. One does not simply walk into Mordor (and show up bright and fresh at the other side).
You need a map for this expedition - and in thesis-terms, that means you need an outline.
Not just a table of contents is what you need, but you need to find out how all the elements can be tied back to your research question. You need something like an "enhanced" table of contents, or a mind-map of your soon-to-be book.
Therefore, I advise you to do the following:
Draw a scheme or diagram of the content, so you know how the different chapters are interrelated
This scheme is not something that you will use in the very beginning - I think it is an excellent element to add such a scheme to the first chapter of your dissertation. In your introduction, you will typically give an overview of what the reader can expect in every single chapter of your dissertation. Go one level up, and present your reader as well how these chapters are logically interrelated by showing the diagram of the contents of your dissertation.
Once you have your overview diagram ready, you might have all the tools you need to get started with your first chapter. While the opinions differ on when you should write your introduction chapter, I think it's not a bad idea to scramble your thoughts together and write a provisional introduction chapter. You can write this chapter just as an exercise in defining the boundaries of what you will discuss in your dissertation, and within which limits you will deal with your research question. Most likely, you will completely revise your first chapter at the end of writing your dissertation, but the first gist of preparing ground for your writing will most likely be kept in there.