In this paper, logical chain is introduced to be the uniform preference representation mean in iterative reasoning of qualitative preference description language LPD (logic preference description). The main contributions include: (1) A new general framework of several common decision making settings is proposed, and LPD is introduced in the framework; (2) LPD is translated into logical chains by constructing logical chains instead of preference operators. Consequently, the preference reasoning in LPD is reduced to several methods of constructing logical chains so that the complex preferences in LPD can be reused in subsequent handlings. Moreover, the conclusion and the prospective are presented in the end.