And I think that ship-related skills (piloting, navigation, engineering, gunnery, etc.) would tend to just lead to dice-rolling instead of decision-making.
More precisely, the problem is that most ship related skills are merely pass/fail maintenance of the ship that only involve dice-rolling when abstracted but offer no interesting decisions to the player.
Even getting past the question of in a cooperative game is the dominate personality allowing the other players to make choices, in a simulation of a ship in combat players assigned to shields or engineering or even gunnery stations are generally only rolling to see if they can do the task in abstract and don't have any interesting choices to make when performing that task. The pilot or captain may be deciding where to move but everyone else is just doing the same thing over and over again.
I have no solution for this, and I think it's relevant to the thread because a party travelling can be abstracted to a ship regardless of the mode of travel. There are "stations" or "roles" in the travel where pass/fail can improve the party's chances of success, but there aren't a lot of choices to make. The navigator doesn't have a lot of choices to make in abstract. Try not to get lost. Try to get unlost once you've become lost. It's not playing a game; it's just performing a task.
One of the most interesting mass combat simulations I ever encountered was in the game "Puzzle Pirates" (of all places). The MMORPG had a lot going for it, but the central game everything was built around was an abstraction of ship combat that involved hundreds of participants in a "crew". The crew had jobs repairing the ship, manning the sails, or loading the cannons. Each crew station was represented by a different casual puzzle to solve (think Tetris or Bejeweled Blitz). Loading cannons meant that when the ship fired a broadside, it had more weight of metal available. Repairing the ship kept it from sinking. Manning the sails delivered more movement choices to the Captain of the ship. When the ship took damage, the crew stations would all get penalties that would make the job harder. At the top of this was the Captain who was the only one however playing the ship game (or even much aware of what was going on). The Captain made all the tactical choices, with the resources he had available being created by the crew's ability to solve puzzles quickly.
If we remove the video game puzzles that the crew were playing, they had no real choices to make and nothing interesting to do. If we abstracted their tasks to dice rolls, then the player in that position as crew would be reduced from playing a game to merely performing a task and observing it with no more input on anything than a player of Candyland.
My concern is that attempts at making travel interesting by having roles and resources to manage tend to just have the same effect. No one is making really any interesting choices. As long as you abstract out "foraging" or "scouting" or "navigation" to just a die roll, the player is merely an observer. But at the same time, it's pretty clear playing out such tasks isn't a game either.