For combinatorial, or permutation, optimization problems,
the solution representation and generation mechanism(s) will
necessarily be problem-specific. For example, in the famous
* traveling salesman problem* (TSP) of finding the shortest cyclical
itinerary visiting N cities, the most obvious representation is a
list of the cities in the order they are to be visited. The solution
generation mechanism(s), or * move set*, must, obviously, be compatible
with the chosen representation. For combinatorial problems, it is
common for the move set to permute a small, randomly chosen, part of
the solution. For example, the move set suggested by Lin [43] for
the TSP makes two types of change:

- a section of the route is removed and replaced by the same cities running in the opposite order, e.g. ; or
- a section of the route is moved (cut and pasted) from one part of the tour to another, e.g. .