To predict equilibrium properties of cyclic alkanes using the Gibbs ensemble Monte Carlo method (GEMC), a new transfer move is proposed to sample the configuration space with efficiency. This new method, called reservoir bias Monte Carlo, consists of adapting the reservoir bias insertion in the grand canonical ensemble, proposed by Errington (Errington, J. Chem. Phys. 1999, 111, 9731), to the Gibbs ensemble. It is coupled with a preinsertion bias to favor transfer in the dense phase. Its application appears more efficient than the adaptation of the configurational bias Monte Carlo in the case of cyclic alkanes (Neubauer; et al. Mol. Phys. 1999, 6, 769). The anisotropic united atoms potential for linear alkanes proposed by Ungerer (J. Chem. Phys., 2000, 112, 5499) has been partially optimized to describe the CH? group of cyclopentane and cyclohexane. The resulting parameters are close to those previously determined for linear alkanes. Simulations have been performed at temperatures ranging from 250 to 600 K for cyclopentane, cyclohexane, and cyclooctane. Equilibrium properties are quite well predicted, and critical properties can be evaluated with a good accuracy. It is concluded that the AUA potential, because of a relevant physical meaning, can be transferred to a large range of cyclic alkanes with good success.