Using Internet of Things (IoT) data analytical techniques may address the difficulty of properly scheduling multiple tasks in a multi-robot-based control system of Green Reconfigurable Manufacturing Systems (GRMS). This work provides a new scheduling methodology for optimizing the production schedule of GRMSs utilizing heuristic approaches such as Gradient Descent (GD), Simulated Annealing (SA), Tabu Search (TS), and Genetic Algorithm (GA). The approach assigns tasks to flexible robots through the flow shop problem, which expands the system’s flexibility, significantly improving efficiency and productivity by reducing the maximum completion time. This study proposes a new hybrid approach called GAGD that combines the global exploration of GA with the local exploitation of GD, providing a more robust optimization procedure that converges faster and finds better solutions than other metaheuristic optimization techniques SA, TS, GA, and GD. By comparing GAGD with other heuristics methods, the authors demonstrate its superiority in optimizing the manufacturing schedule of GRMS.