052D has two GT and two diesels, 055 has four GTs. Spruance, ABs and Ticos all have four GTs, and there are a huge number of these ships made. Russians also built the Udalois with six GTs, 2 bigger ones and 4 smaller ones. Cost is higher with 4 GTs but that won't stop you from making a lot of them. Your bottleneck is how fast you can produce the GTs. Note the future Type 054B may also further consume GT production if a ship has one or two of them.
The greater cost of the ship might be on the radars --- all those AESAs on the 055 add up to a large bill. 052D still features a good amount of mechanical radars. Something like the 055 might be price impossible if done in the West; the US took out the S-band AESA SPY-4 radars from the Zumwalt --- the functional analog to the 055's main Type 346B sets --- and which are originally part of the Zumwalt's sensor set, to save on costs.
Going forward however, spectrum dominance by ECM/ESM will be more and more important, AESAs can make the emissions harder to detect and harder to jam with better abilities to detect and target stealthier objects, so I expect the 055 to pay off more than the 052D in the long run, while something completely based on mechanical radars like the 054A and 056 can quickly become obsolete unless they are MLU'ed with a better set of radars.