The present paper aims to develop a new Water Quality Index (WQI) based on Data Envelopment Analysis (DEA). Rather than using subjective weights of a judgmental process as inputs of the DEA model, we propose more objective variables, identified as “optimistic closeness values”, appropriately derived from the observed values of the hydrochemical parameters.

The proposed approach was employed to assess the water quality of 47 dams in Algeria, defined with a dataset of 10 hydrochemical parameters. The results of the DEA-based WQI application revealed that (i) 21.27%, 27.66%, 25.53%, 4.25% and 21.27% of the total dams are categorized as “Poor”, “Marginal”, “Medium”, “Good” and “Excellent” water quality, respectively; (ii) the best water quality is found in “Kissir” and the worst one in “Bougara”; (iii) a priority scale on the hydrochemical parameters can be set for the treatment of water using the notion of slack value.

Collectively, the new methodology has proven its effectiveness not only for categorizing or ranking sites based on water quality but also as an alternative tool to be used to assist decision-makers in allocating funds and managing water resources.

本文旨在基于数据包络分析（DEA）开发一种新的水质指数（WQI）。我们不是将判断过程的主观权重用作DEA模型的输入，而是提出了更多客观变量，这些变量被识别为“最佳接近度值”，可以从水化学参数的观测值中适当得出。

该提议的方法被用来评估阿尔及利亚47个大坝的水质，该水坝由10个水化学参数的数据集定义。基于DEA的WQI应用程序的结果显示：（i）总水坝的21.27％，27.66％，25.53％，4.25％和21.27％分为“差”，“边际”，“中等”，“良好”。和“优秀”水质；（ii）在“基西尔”发现最好的水质，在“布格拉”发现最差的水质；（iii）可以使用松弛值的概念为水处理设置水化学参数的优先级。

总的来说，新方法论证明了其有效性，不仅可以根据水质对站点进行分类或排名，而且还可以用作辅助决策者分配资金和管理水资源的替代工具。