06-22-2014, 08:49 AM
For adjusting production based on demand, you don't necessarily need you predict the future, just the past and present. Based on last month's performance, you can scale down production for next month. Not perfect but that's what I do myself anyway. You can do it either proportionally or even implement a simple PID algorithm, maybe use it for harder AI and when your Sales/Marketing get better trained or something. Maybe as a research?
If not that, you could greatly simply dealing with production by either making a regional production manager so you don't have to hunt down models factory by factory (try having 12 different models and having to cycle through 5-6 factories sliding things up or down every few turns, it's just mind numbingly tedious and it doesn't need to be. Maybe make it manual for the first few years/decades but be able to be automated on easy/researched on hard?) or make it so factories contribute to a total regional production capacity in production units, i.e. NA/SA/EUR/Asia etc... and control production in regions.
Another cool idea/feature would be "market research" or focus testing or whatever you want to call it. Right now whenever I design/release new vehicles, I'm pretty much taking a stab in the dark, adjusting wildly until I find a decent price, then slowly depreciating it by hand over many models at once and it gets very tiring very quickly. My idea would be to either fund market research or include it into the vehicle cost via the demographic focus research or whatever it's called that doesn't seem to do anything I can tell. The more you pay for, the more "data" you get, showing you roughly how much your vehicle would sell at what price points with however much certainty. Implementation would seem rather simple, you can just run separate simulations at different price points and print out roughly how much you'd sell at different prices, etc... and you can take it however far you want, i.e. add info about who's buying in what demographic.
If not that, you could greatly simply dealing with production by either making a regional production manager so you don't have to hunt down models factory by factory (try having 12 different models and having to cycle through 5-6 factories sliding things up or down every few turns, it's just mind numbingly tedious and it doesn't need to be. Maybe make it manual for the first few years/decades but be able to be automated on easy/researched on hard?) or make it so factories contribute to a total regional production capacity in production units, i.e. NA/SA/EUR/Asia etc... and control production in regions.
Another cool idea/feature would be "market research" or focus testing or whatever you want to call it. Right now whenever I design/release new vehicles, I'm pretty much taking a stab in the dark, adjusting wildly until I find a decent price, then slowly depreciating it by hand over many models at once and it gets very tiring very quickly. My idea would be to either fund market research or include it into the vehicle cost via the demographic focus research or whatever it's called that doesn't seem to do anything I can tell. The more you pay for, the more "data" you get, showing you roughly how much your vehicle would sell at what price points with however much certainty. Implementation would seem rather simple, you can just run separate simulations at different price points and print out roughly how much you'd sell at different prices, etc... and you can take it however far you want, i.e. add info about who's buying in what demographic.