Over the last two decades, many modeling and optimization techniques have been developed for earth observation satellite (EOS) scheduling problems, but few of them show good generality to be engineered in real-world applications. This study proposes a general modeling and optimization technique for common and real-world EOS scheduling cases; it includes a decoupled framework, a general modeling method, and an easy-to-use algorithm library. In this technique, a framework that decouples the modeling, constraints, and optimization of EOS scheduling problems is built. With this framework, the EOS scheduling problems are appropriately modeled in a general manner, where the executable opportunity, another format of the well-known visible time window per EOS operation, is viewed as a selectable resource to be optimized. On this basis, 10 types of optimization algorithms, such as Tabu search and genetic algorithm, and a parallel competitive memetic algorithm, are developed. For simplified EOS scheduling problems, the proposed technique shows better performance in applicability and effectiveness than the state-of-the-art algorithms. In addition, a complicatedly constrained real-world benchmark exampled by a four-EOS Chinese commercial constellation is provided, and the technique is qualified and outperforms the in-use scheduling system by more than 50%.