Wireless communication technology drives the cyber-physical integration of digital twins in smart manufacturing. Orthogonal frequency-division multiple access (OFDMA), a promising multi-user transmission mode, can efficiently support the transmission of small frames to a group of sensors simultaneously. However, in time-sensitive scenarios, ensuring real-time data connectivity and monitoring remains a key challenge for OFDMA applications in smart factories utilizing digital twins. In this paper, we propose an OFDMA-based time-triggered scheduling strategy for low-latency communication in wireless sensor networks deployed in digital twin-enabled smart factories. Conflict-free scheduling constraint models based on integer programming (IP) are established, including latency constraints, RU continuity constraints, and TTI-RU constraints, for the joint scheduling of transmission time intervals and resource units. We further abstract the two-dimensional scheduling problem into a one-dimensional scheduling problem of time-frequency resource blocks (RBs), which decreases the number of constraints and expands the scheduling scale within limited time. We apply IP to design a scheduling method with minimal total end-to-end latency as the optimization goal. The experimental results demonstrate the advantages of RB-based scheduling in terms of latency and scheduling scale.