Abstract: A single machine scheduling problem with family setup time was studied to solve the real-life scheduling problem in a steel wire factory. According to the requirements of the real problem, the maximum lateness was minimized. As the problem is NP-hard, a heuristic algorithm was proposed to solve the problem. The problem's characteristics were introduced into the algorithm to improve the search efficiency and running time. The proposed algorithm was tested both on randomly generated problems and on real-life scheduling problems from the workshop. The results show that the proposed algorithm can obtain a near optimal solution in a short time.