The robot-task-sequencing planning problem is investigated in this paper, where multi-robot tasks with resource conflicts and precedence constraints are involved making the problem challenging and complex. An effective approach is proposed to minimize the production cycle time while handling the high complexity effectively. The approach decouples the entire problem into simpler ones and then solves them separately to obtain the final solution. Firstly, given the multi-robot tasks, how to coordinate the operation of each single robot is determined. This is achieved by searching the shortest path in the graph where the tasks are organized in the optimal order without the consideration of resource constraints. Then, the resource allocation is modeled as a constrained assignment problem with the objective of minimal cycle time, which is solved with graph optimization. Consequently, the best-fit robot configurations for task execution are selected according to a heuristic optimality metric, based on which the task execution sequence obtained in previous steps is re-optimized. Finally, the robots involved in the task are regarded as a whole to deduce feasible trajectories. Experimental results demonstrate that the proposed approach can generate solutions where the robot can complete the given tasks effectively with feasible scheduling. Moreover, the proposed approach is successfully implemented in a pragmatic system which tests signals at both ends of components in a circuit board.