There is a space empire consisting of 100 planets, these planets generate 1,000 recruits that can be used as ships CREW.

You build SHIPS, there is no limit on number of constructed ships but the number of ACTIVE SHIPS (fleet) is limited by crew available.

Now, the question is how to handle the crew and ships relationship?

Typically in such games crew is used as a hard limit for building ships (you can't build more ships that crew allows) but I can't do it, I need a pool of obsolete/unused ships that are not manned at the moment. My system is supposed to be highly automated (the AI manages which ships are active and which are waiting in reserve, send them to repairs and move older ships to reserve when better ones are built), that part is easy.

But what to do if:

- a ship is destoyed (so the crew capacity is freed) - should the crew immediatelly man the next best unmanned ship OR is the crew "dead" and you need to wait till the planets regenerate their pool of recruits?

- a ship is damaged and is sent for repairs - should the crew be able to just switch to a fully operational unmanned ship from reserve (which could be quite unbalancing in case you have a large pool of older unmanned ships, it could make wars quite long; also what to do if their original (better) ship is repaired, should they switch back again)?

I suggest a realistic simulation: less complications and arbitrary decisions for you, and less surprises and aberrant strategies for the players. Abstractions (like "planets regenerating their pool of recruits" or "switching to a ship from reserve") are a threat to the integrity of your game.

• Damaged ships can be reclassified for automation purposes (e.g. large ship with disabled weapons, capable of transporting cargo but in need of some escort) but they are still flying: sending them to a shipyard for repairs or major upgrades is an order like any other.
• Reduced-crew ships (e.g. out of redshirts after too many planetary operations) are likewise operating as usual, with reduced capabilities; sending them where new personnel can join the crew should be the player's choice.
• Severe cases of damage (e.g. no life support) and crew reduction (too few people to pilot) should be treated as a lost ship. Other ships can rescue crew and other stuff, particularly in the case of inoperative engines.
• The crew of destroyed ships can leave in escape pods or the like, which you can treat as other ships or, more simply, as people reappearing in a preassigned or AI-selected destination after a predictable time (with some chance of dying instead of arriving).
• Crew assigned to other ships has to travel explicitly (on board a suitable ship) to a base or to a rendez-vous in space, with normal travel time (not "immediately"). Starting such travel on board a damaged ship in the case of recycled crew or on a planet in the case of newly trained officers is a minor detail.
• Every ship is located somewhere, including mothballed ones. It takes a crew, either directly assigned or manning a big carrier ship, to move "reserve" ships where a full crew will make them operational again. Is there a large difference of quality and/or quantity between the minimum crew required to fly around and the full crew required for war activities?
• The number of ship officers coming from a certain planet can be limited by the rate of training new starship personnel (e.g. each academy begins a 40 cadets five years course every six months, with 75% dropouts: 10 officers every six months at full capacity) and/or by the rate of breeding exceptional people (e.g. if one child in a million has the required psychic powers to be a warp pilot, a population of 1 billion women who have an average of a child every 50 years yields 20 potential warp pilots per year). Neither limit gives rise to a hard cap on personnel coming from a certain planet, as their numbers, their social impact and their cost of living are negligible.
• Moreover, anything that happens on the war front has no impact on training replacements for lost crews: qualified starship personnel is supposed to be trained as fast as possible at all times, especially during a war.
• Both spare ships and spare personnel can be, and should be, shipped around planets and bases according to strategic predictions: places near the front are appropriate for gathering new officers, new ships, repair parts, and ships in need of repairs and crew.

You can effectively limit available personnel, and active fleet sizes as a natural consequence, by making the duration of a war much shorter than the time needed to get significant numbers of new officers.

If a ship can reach any place on the map in 2 days, and there are (at the beginning) 9 mothballed ships for each active one, about 40 days (slightly more in case of losses) are enough to fly every ship you have into battle. Meanwhile, the number of new officers produced is negligible compared to the merely shuttled around ones.

Things I can't do:

- crew being assigned by player's command - there would be 1,000 ships in a quite early stage of the game, it would be unmanageable; plus it would be inconsistent with the rest of the game. Everything regarding this needs to be fully automated (well, maybe except setting some overall global priorities to your fleet behaviour, like when the AI should retreat with damaged ships, but that's still just a variable of the automated system, not an order issued by the player)

- a realistic simulation - I have no budget (development time) for this :) I need it reasonably simple and easy to implement.

More about the overall feel of the game for which I need this subsystem is here: http://www.gamedev.net/topic/644645-sectors-battles-strategy/

As far as saving crew goes I would first look to your actual battle model. Just saying "The ship has escape pods, and X% of crew magically teleports back to the home world for reassignment" sounds kind of dull to me. Instead having various levels of ship damage (ie, weapon of mass destruction impact on main reactors causing the whole ship to go up vs the hull getting chewed by small weapons fire to the point it is no longer deemed spaceworthy) come into play, as well as the results of the engagement (Your fleet routs and retreats vs a draw with both sides pulling out vs a resounding victory) being the deciding factor in how many actually get to return.

Options like a focus on Search and Rescue tech to recover escape pods in battle, or intelligence and special operations for recovering POWs, could really make a difference in a game that fell into total war.

As for crew, I suggest limits on logistical movements. Have the player sort out trade/convoy routes which have a set speed and volume, and let the AI move crews around. Have it prompt you if it detects shortages and such so you can quickly spot if you need to enable an emergency convoy or something to move additional forces around. Planets then become resource pools and sinks.

It might be normal that there is no balance between crew vs ships.

Another thing is that crew (based on population) will be rather static (birth rate is not that big), unlike ships (produced each turn).

Generally, I want the player to create formations and ships will be mere weapons (just like in real military). You create a combat squadron (formation) that is supposed to consist of 3 recon ships, 6 combat ships and 2 support ships + the respective crew and the game AI will try to auto assign the best ships to a newly created formation. Note that you can create a formation even if you don't have enough ships for it yet.

As far as saving crew goes I would first look to your actual battle model.

Here is the battle model description: http://www.gamedev.net/topic/644645-sectors-battles-strategy/#entry5072337

As for crew, I suggest limits on logistical movements. Have the player sort out trade/convoy routes which have a set speed and volume, and let the AI move crews around. Have it prompt you if it detects shortages and such so you can quickly spot if you need to enable an emergency convoy or something to move additional forces around. Planets then become resource pools and sinks.

The player is an emperor (very high level of command) and he won't be able to alter any trade/convoy routes. It will be all either automatized or abstracted.

The only logistics related decision would be:

- creation of a starbase/supply depot in a certain sector (which will help with all transport in that sector)

- what percentage of industry is set on building new freighters

- what percentage of industry is set on building new convoy escort ships

- making trade agreements with other empires

Well the emperor, if he is to have any control over military strategy, should still be able to say "I want a stockpile/troop depot HERE, make it so", and have the logistics system automate things and pull suitable resources from else where, and divert resources accordingly. If you have a weak freighter infrastructure then any kind of reorganization of your troops/ships would be delayed. If you keep your investment strong then you will have fewer delays (but will likely have fewer ships and trained staff to work with)