Record Display for the EPA National Library Catalog


Main Title Using a neural network to estimate solvent consumption
Author Capone, R. L. ; Chappell, P. J. ;
Other Authors
Author Title of a Work
Chappell, P. Jeff.
Capone, Ronald L.
CORP Author Capone (Ronald L.) and Associates, Arlington, VA.;Environmental Protection Agency, Research Triangle Park, NC. Air and Energy Engineering Research Lab.
Publisher U.S. Environmental Protection Agency, Office of Research and Development, Air and Energy Engineering Research Laboratory,
Year Published 1993
Report Number PB93-173110 ; EPA/600/A-93/063 ; EPA P.O. 2D1328NATA ; AEERL-P-1001
Stock Number PB93-173110
OCLC Number 39877052
Additional Subjects Neural networks ; Solvents ; Study estimates ; Pollutants ; Environmental surveys ; Forecasting ; Graphic arts ; Printing inks ; Economic impact ; Productivity ; SIC 27
Library Call Number Additional Info Location Last
EKBD  EPA-600/A-93/063 Research Triangle Park Library/RTP, NC 09/25/1998
NTIS  PB93-173110 Some EPA libraries have a fiche copy filed under the call number shown. 07/26/2022
Collation 4 p. : ill. ; 28 cm.
The paper discusses a neural network, using the backpropagation paradigm, that is taught the relationship between employment in the graphic arts industry--(Standard Industrial Classification Code (SIC) 27)--and economic variables and solvent consumption by SIC 27. The project is a proof of concept whose objective is to a relationship using national-level data, which are known, and apply it to estimating solvent consumption on the county level, where data are thus far not available. The network accurately learns a relationship from national data. Although definitive testing is not yet possible due to data limitations, there are indications that the national relationship can be used to estimate county-level solvent consumption. Network inputs are SIC 27 employment, productivity for the current and 1 prior year, and an eight-element 'signature' of quarterly economic changes in output from non-durable industries. One hidden layer of two processing elements connects the 11-element input layer to a 1-element output layer. NeuralWare Professional II Plus Version 4.0 was used as the platform. Training requires 30,000 iterations and results in a Pearson's r value of 0.99. The best result achieved by ordinary least squares regression was 0.93.
Cover verso title. EPA Project Officer: P. Jeff Chappell. "EPA/600/A-93/063." "PB93-173110."