Configurational-bias-vaporization Monte Carlo simulation method developed previously has been used to simulate phase equilibria of AB random copolymer solutions based on a lattice model. Random copolymers are modeled as linear chains composed of segment A and segment B, for which the numbers of segments of them are r(A) and r(B), respectively. Coexistence curves of random copolymer solutions with total chain length r = r(A) + r(B) up to 200 and with chain composition varied are presented. Different sets of energy parameters are used. It is found that the critical density rho(c) is mainly determined by the total chain length I, while the reduced critical temperature T-c* is determined by both the total chain length r and the chain composition f. Similar to Kambour et al. and Brinke et al.＇s work, an effective interchange energy parameter epsilon(eff) is introduced to treat the simulation results. Here epsilon(eff) is a function of chain composition and various interchange energy parameters. With this measure, coexistence data of random copolymer solutions with different chain compositions can be mapped into a single curve for most cases. Phase behavior of these random copolymer solutions can then be predicted by coexistence curves of corresponding homopolymer solutions. However, the results are less satisfactory when monomers A and B are strongly repulsive toward each other.