Abstract: The manufacturing flow of steelmaking and continuous casting is a complex multiple-phase and multiple-product production process.The production scheduling problem in this manufacturing flow can be seen as a job shop scheduling problem.An improved genetic algorithm for solving this problem was proposed and the improved aspects were as follows:rank-based fitness assignment,job filter order-based crossover operator,and mutation rate according to an exponential function relation.Twenty-four benchmarks were comparatively investigated and the result shows that the improved genetic algorithm has a better capacity of seeking optimum than a traditional genetic algorithm.The production scheduling problem of steelmaking and continuous casting with sixteen plans and six procedures was computed using the improved genetic algorithm.It is shown that the algorithm is effective.