I'm kind of bemused by your specialisms, but variety is the spice of
life and all that.
I am not an expert in the fields you list so probably am not well placed
to give good examples. Also the examples will depend on the granularity
at which you want to model things and whether it is the day-to-day
operation that is being modelled or the actual creation of plant.
There may of course not be good examples in those industries.
In any case, the last two examples carry across. If you were installing
ovens you might monitor their operation closely for a fixed period after
installing them. Similarly, any project-based activity might have a
management activity associated with it as a fixed overhead.
The finish-to-finish example must also carry across if you replace
review with inspection/sign-off and use any design process.
I would have thought that the start-to-start example would also carry
across simply. If you were making a batch of ovens/cars and had to
start testing at some point, when can the testing actually start?
You should note that finish-to-finish is by far and away the most common
link type. It probably accounts for 99% of all links.
You should also note that the plan is a model of your development path
and to a certain extent the link types you use depend on your view of
the process (for example in the review example below, you could quite
easily take the view that the review cycle starts as soon as the design
phase starts so the start of the design drives the start of the review).
Generally speaking you may well find that trying to use the other link
types will complicate things unduly and that there is merit in keeping
your model as simple as possible.
