中文摘要：

试卷自动生成是一个多目标的问题，遗传算法可以搜索全局最优解，将遗传算法用于自动组卷算法中，并引入平衡算子和权重算子计算算法的概念，有效地解决难以寻求最优解的问题．实验结果表明：该算法有效地避免了通过一些常规的方法所造成的弊端，并且提高了搜索全局最优解的能力，加快了收敛速度．该方法不仅在教育领域有一定的实用性，也具有潜在的应用价值．

英文摘要：

Considering the problem on generating test papers is multi-objective and the genetic algorithm（GA） can search the globally optimal solution, a GA-based algorithm for test paper generation automatically is presented, and effectively solves the problem that it is hard to find the optimal solution, by introducing the concepts of the equilibri- um operator and the weight operator. The experiments show that this algorithm effectively avoids the disadvantages caused by some conventional methods, and moreover, it improves the ability of searching a globally optimal solution and increases the convergent speed. This method not only is effective for the kind of practical problems in education field, but also has potential applications in several problems in engineering.