Funny you mention that, because last night we did some more testing. We literally blew up every building in Syrta with C4 and mortars and turned the town into a moonscape and were still unable to cap.
If I may offer some ideas, maybe requiring the death of every AI generated in that zone isn't the best strategy when it comes to capture. Because who knows if these AI get stuck in the ground or wander off somewhere. My friends and I had a few ideas.
I have only a cursory knowledge of arma scripting, so some of these ideas may be difficult to implement:
Some sort of function to check whether the number of enemy forces left is below a certain threshold, or percentage, and if yes the area is captured. The idea here is that you don't have to exterminate the area to the last man, but do significant enough damage to secure the area.
Make only, and only the AI currently in the red capture zones count towards the area still being held by the enemy. I imagine this would be difficult to implement, solely because when the AI's are generated, I gather you might have it set to a total number of people that need to be neutralized in order for the area to be considered captured. And having some kind of function to change an AI or vehicle not count or count towards the zone would be more difficult than it's worth.
Some type of object or area that you have to seize. There seems to be a lot of importance on flags in this mission, and in my mind perhaps you could place a flagpole down in the center of town and create some function to initiate an animation that would be considered taking down the enemy flag and replacing it with the FIA one. The player character, or FIA forces that start the capture of this flag could perhaps trigger another function where the AI would start to mobilize towards the town center, and flag to try to stop you. This could create some auxiliary gameplay by allowing you to implement wandering CSAT forces who will attempt to recapture areas you've previously secured, creating a more dynamic feel to the map.
Of course if you can get the regular capture method working smoothly that would be acceptable. I just feel like the need to neutralize every enemy in an area bogs down the map a bit. It can literally take hours to capture each area, and in a public server setting, people generally don't have the patience to commit an entire day to capturing the entire map. I feel like this map has a lot of potential, and I also don't mean to try to distort your vision for where you want the map the go. I'm just offering some observations.