This paper addresses the fixed-time consensus issue in uncertain multi-agent systems (MAS). It introduces novel estimation methods for the maximum settling time, which considers all controller components but may also have less conservatism than traditional methods. Based on these results, a distributed control scheme and the corresponding consensus criterion are proposed, enabling MAS to achieve a fixed-time consensus and provide a more accurate estimation of the maximum settling time than the traditional method. Additionally, a distributed intermittent control scheme is also provided to enable the MAS to solve the issue of practical fixed-time consensus and be controlled only in a series of disjoint periods. Moreover, considering the case where the MAS has no external disturbance, another distributed intermittent control scheme is introduced to achieve fixed-time consensus under a directed topology. Finally, the results are supported by numerical examples.