Because of the complexity and importance of the project bid evaluation, this study primarily utilizes DEA method to select reasonable and effective bidding scheme. By combining with genetic algorithm and BP neural network algorithm, this text eventually presents an approach of using real-coded genetic algorithm with adaptive mutation to train the weights of the BP neural network. On the basis of DEA results, the network is tested and adjusted; the second bid evaluation eventually are obtained; sorting and optimization of bid scheme are realized; the best bid winner is eventually determined. This method avoids evaluation errors that are brought by the subjectivity and orientation of traditional expert evaluation, greatly improves the objectivity of the evaluation and the work efficiency, and is applicable to other types of project evaluation and decision.%考虑建设工程招投标中评标环节的复杂性和重要性，本文利用DEA方法筛选出初步合理有效的投标方案，并结合遗传算法和BP神经网络算法，提出了用实数编码的自适应变异遗传算法训练BP神经网络权重的混合算法。依据DEA初评的结果进行网络的测试、调整，最终对各投标方案做出二次评价，实现投标方案的排序与选优，决策出最佳的中标单位。该方法避免了传统专家评标的主观性和倾向性带来的评标误差，大大提高了评标的客观性和工作效率，并适用于其他类型项目的评价和决策。
展开▼
