基于不确定信息条件下的多属性群组决策评价模型

Multi-attributive group decision evaluation model under the condition of uncertain information

建筑安全管理的复杂性要求将决策者的知识和经验有效结合以利于做出正确的决策判断.逼近理想解排序方法(TOPSIS)作为一种实用且有效地处理多属性决策问题的方法被广泛地应用到建设管理领域.将该方法与灰数理论相结合,通过建立不确定信息条件下的多属性群组决策模型来进行建筑安全绩效评价.首先采用区间型灰数矩阵描述专家评价信息的不确定性;其次引入线性变换函数对灰数评价矩阵进行规范化处理以改善评价信息偏差问题,并采用明可夫斯基距离函数来克服传统TOPSIS方法中初始权重影响过大的缺陷;最后结合建筑安全绩效评价实例证明该模型算法的科学性和有效性,并易于拓展到类似决策评价中.

Owing to the complexity in the construction safety management, integrating expert＇s knowledge and experiences to make appropriate decisions is necessary. TOPSIS （technique for order performance by similarity to ideal solution） is a practical and useful technique in dealing with multi-attribute decision making problems,, and has been widely employed in construction management. Combined with traditional TOPSIS and grey number theory, the article proposes a multi-attribute group decision making model under the condition of uncertain information in the construction safety performance evaluation. Firstly, interval grey number evaluation matrix is applied to describe the uncertain decision information by expert. Secondly, a linear transformation function is adopted to construct the normalized grey decision matrix to avoid the information risk. Besides, Minkowski distance function is further integrated to overcome the over effects of weighing in the original TOPSIS technique. Finally, an application example shows that the proposed model is reasonable and efficient, and can easily extend to similar decision problems.