酒店用水量影响因素与预测模型研究

Study on factor analysis and prediction model of water consumption in hotel

基于实地走访和问卷调查,收集了全国共计18个酒店的用水量及相关信息数据。采用相关性分析确定了与用水量显著相关的影响因素,并依据数据特点,选择线性回归和稳健回归构建了不同的用水量预测模型。利用各模型预测3栋未包含于样本数据集的酒店的用水量,从中选出预测能力最佳的模型,并评估其是否具有实用价值。结果表明,稳健回归模型的预测能力最佳,其预测值与真实值的相对误差在25%以内,并且与依据相关标准计算出的用水量相比,稳健回归模型的预测值更接近真实值。

Water consumption and related information data of 18 hotels across the country were collected by field research and questionnaire. The factors significantly related to water consumption were determined by correlation analysis. According to the data characteristics, linear regression and robust regression were selected to establish two water consumption prediction models. Using each model to predict water consumption of 3 hotels which are not included in the sample data set in order to select the model with the best prediction ability and evaluate whether it has practical value. The results showed that the prediction ability of robust regression model was the best, whose relative error was within 25%. Moreover, compared with the water consumption calculated by relevant standards, the predicted values of robust regression model were closer to the real values.