基于间隔率和遗传算法的多资源均衡优化研究

为解决传统编码方式易产生无效解的问题，提出基于间隔率的编码与解码方法，从而避免在交叉、变异等遗传操作中出现违反作业间逻辑关系的情况。由于方差更能体现项目在各时刻资源需求的不均衡程度，为此建立了以资源方差最小为目标函数的多资源均衡优化模型。结合案例，详细叙述了该算法的设计参数和应用步骤，结果表明，该方法较传统编码方式具有收敛速度快、优化效果明显等优点，从而验证了该方法的可行性和有效性。

Leveling Optimization of Multiple Resources Based on Interval Rate and Genetic Algorithm

The traditional encoding always generates invalid solution. An encoding and decoding method based on interval rate was proposed to solve the problem. The method avoids the violation of the logical relationships between jobs in crossover op- eration and mutation operation. A multi - resource leveling optimization model was analyzed and designed, whose objective func- tion was resource variance minimum, because of the variance better reflecting the uneven degree of the resource requirements. Fi- nally, a case detailed the design parameters and application steps of the algorithm, and was used to verify the feasibility and ef- fectiveness of the method. The study indicates that the method is better than the traditional encoding. The convergence speed is faster and optimization effect is more obvious than that of the traditional one.