To achieve URLLC in an IEEE 802.11ax-based relay-assisted industrial network, we propose four substantial processes: cluster formation, optimal route selection, priority aware scheduling, and optimal antenna selection. Initially, the network of relay nodes is segregated into multiple clusters via a weighted spectral clustering with optimal cluster size algorithm. Within each cluster, intra-cluster routing is performed using a game theory-based optimal matching (GTOM) algorithm. In the GTOM, players (forwarder candidates) engage in a one-to-one matching game, and each player is evaluated according to the residual energy, link quality, and delay. Aggregated data are organized for transmission by the cluster head (CH), with two phase priority-aware medium access control (TP-Prior MAC) scheduling. To optimize the data transmission between the CH and the base station, bacterial foraging optimization algorithm based antenna selection is employed. Extensive simulations confirms that the TP-Prior MAC method achieves 50% better performance with regard to the URLLC latency (end to end). Additionally, the proposed method exhibits a better impact on the throughput, packet delivery ratio, packet loss rate, and energy consumption in an industrial wireless network.