In a search for new and improved methods for accurately predicting noise levels for all situations, two new prediction methods (the NCHRP 3-7/3 Revised Design Guide, RDG, and Ontario Highway Noise Prediction Method) were evaluated and compared with actual field measurements and the results using HUSH (a computer program based on the "Michigan Model" computerized version of NCHRP 11m/144). The RGD is a physical model designed for freely flowing traffic manifested in both a nomograph and computer program; and the Ontario method is a regression line model. in the form of nomographs based on 133 noise measurements at 120 locations near rural and urban freeways. Results for comparison purposes were obtained by inputting the actual traffic and geometric parameters into the 3 models, and comparing the outprints to the field measurements. The results show the HUSH overpredicts by 3 to 4 dBA for most situations, and that the Ontario method appears to the most accurate in terms of mean of the difference. The RDG predicts accurately with low standard deviations for all situations for the L10 descriptor; as distance increases, RDG L10 starts overpredicting. The results indicate that the simple relationship for (L10 - L sub dq) = 3 may apply for most practical situations. Also, there appears to be a range in volume below which the (L10 - L sub eq) relationship complicates, or the L10 descriptor becomes unstable.

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00131402
  • Record Type: Publication
  • Files: TRIS, TRB
  • Created Date: May 14 1976 12:00AM