The parallel-machine scheduling problem has been an active research field over the past decades because of its practical applications. The present study proposes job-driven scheduling heuristic (JDSH) and machine-driven scheduling heuristic (MDSH) to solve a practical-size dynamic parallel-machine scheduling problem with stochastic failures. The proposed methodology is applicable for solving an identical, uniform or an unrelated parallel-machine scheduling problem. In addition, it is sufficiently robust to accommodate product mix or demand changes. A practical-size wire-bonding workstation from an integrated-circuit packaging plant is adopted for the empirical application. The empirical results are promising, and lead to the use of MDSH as the scheduler a priori. The proposed heuristics are suitable for practical application due to their efficiency and effectiveness. When an automatic shop-floor control system is available, the proposed heuristics are seen to be superior schedulers for providing a real-time scheduling decision.