公路项目安全性评价费用测算方法研究

为了科学地测算公路项目安全性评价费用、规范评价市场价格以及保证评价报告质量,以2016—2019年的高速公路安全性评价项目数据为基础,分析项目的投资额、建设里程和安全性评价费用之间的关系,运用规范变换原理,通过幂函数和对数变换组成的规范变换式,将投资额、建设里程等效于一个线性化的规范因子,建立了基于规范变换的一元线性回归模型（NV-ULR）,并对模型进行了验证。结果表明,模型可用于测算投资额在20~200亿元之间或建设里程在20~180km之间的安全性评价费用,模型测算的相对误差基本能控制在10%以内,且模型精度高于常规一元线性回归模型。该测算方法可应用于公路项目安全性评价取费标准制定。

Calculation Method of Service Fee for Highway Safety Audits

In order to scientifically calculate service fees, offer fair market prices and ensure the quality of reports of highway safety audits, the average prices were surveyed from 2016 to 2019. The relationship among the investment amount, construction mileage and fees for highway safety audits was analyzed. Based on theory of canonical transformation, power function and logarithmic function were applied to the normalized transformation. Thus, the two factors could be simplified and generated an "equivalent" linear canonical factor. The Univariate Linear Regression of Normalized Variable(NV-ULR) prediction model was built and verified. The result shows that the NV-ULR prediction model can be used to measure service fees for highway safety audits, while the highway project investment ranges from 20 billion yuan to 200 billion yuan or the construction mileage ranges from 20km to 180km.The relative error of the NV-ULR prediction model is within 10%, which is higher than the accuracy of the conventional univariate linear regression model. This method can be applied to formulate the standard of service fee for highway safety audits.