In steelmaking-continuous casting production process, there are complex physical and chemical changes, and processing time presents strong volatility. The scheduling scheme based on standard processing time is difficult to satisfy the actual production needs. Aiming at steelmaking-continuous casting scheduling with uncertain processing times, the robust optimization method is used and the combined box and polyhedral uncertain set is constructed to deal with uncertainty. A robust optimization scheduling model is established with the objective of minimizing the total waiting time. Aiming at its characteristics of two different types of decision variables, namely, discrete assignment variables and continuous robust equivalent transformation variables, a hybrid encoding genetic algorithm is proposed. A parallel evolution strategy is proposed for the hybrid coding characteristics of chromosomes. To improve its search ability, the elitism strategy and adaptive parameter are introduced. The simulation experiments are designed based on the actual production data of various scales, and the results show the effectiveness of the proposed model and algorithm.