In this paper,a utility-based feedback delay-aware and buffer status-aware( FABA) scheduling scheme is proposed for downlink multiuser multiple-input multiple-output orthogonal frequency-division multiple-access( MIMO-OFDMA) systems. The FABA scheme allocates subcarriers to multiusers with an objective of not only maximizing the total system capacity but reducing the system packet loss rate as well. We design a utility function which consists of a feedback estimate module,a proportional fairness module and a buffer monitoring module. The feedback estimate module is used to improve the system throughput by utilizing the Automatic Repeat-reQuest( ARQ) feedback information to combat the fast time-varying fading condition. The proportional fairness module can guarantee the scheduling fairness among users,and the buffer monitoring module can utilize the transmitting buffer status information to avoid high packet loss rate of the system caused by the system congestion. The FABA scheme then formulates the scheduling problem into a problem of overall system utility maximization. We solve the problem by using a heuristic algorithm with low computational complexity. The simulation results show that the proposed FABA scheme outperforms the existing algorithms in terms of the system throughput and the packet loss rate and can also guarantee the fairness demand among users. In this paper, a utility-based feedback delay-aware and buffer status-aware (FABA) scheduling scheme is proposed for downlink multiuser multiple-input multiple-output orthogonal frequency-division multiple-access allocates subcarriers to multiusers with an objective of not only maximizing the total system capacity but reducing the system packet loss rate as well. estimate module is used to improve the system throughput by utilizing the Automatic Repeat-reQuest (ARQ) feedback information to combat the fast time-varying fading condition. The proportional fairness module can guarantee the scheduling fairness among users, and the buffer monitoring module can utilize the transmitting buffer status information to avoid high packet loss rate of the system caused by the system congestion. The FABA scheme then f ormulates the scheduling problem into a problem of overall system utility maximization. We solve the problem by using a heuristic algorithm with low computational complexity. The simulation results show that the proposed FABA scheme outperforms the existing algorithms in terms of the system throughput and the packet loss rate and can also guarantee the fairness demand among users.