Abstract
This paper compares Genetic Programming and the Classification and Regression Trees algorithm as data driven modelling techniques on a case study in the ferrous metals and steel industry in South Africa. These industries are responsible for vast amounts of greenhouse gas production, and greenhouse gas emission reduction incentives exist that can fund these abatement technologies. Genetic Programming is used to derive pure classification rule sets, and to derive a regression model used for classification, and both these results are compared to the results obtained by decision trees, regarding accuracy and human interpretability. Considering the overall simplicity of the rule set obtained by Genetic Programming, and the fact that its accuracy was not surpassed by any of the other methods, we consider it to be the best approach, and highlight the advantages of using a rule based classification system. We conclude that Genetic Programming can potentially be used as a process model that reduces greenhouse gas production.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Lotz, M., Brent, A., Steyn, H.: Addressing the need for a Clean Development Mechanism (CDM) specific project management strategy. South African Journal of Environmental Management Science NS 12(2), 228–241 (2009)
Intergovernmental Panel on Climate Change (2009), http://www.ipcc.ch/
United Nation Framework Convention on Climate Change (UNFCCC), http://cdm.unfccc.int
Lotz, M., Brent, A.: Towards a comprehensive Clean Development Mechanism (CDM) approach for biodiesel. Energy and Sustainability II, 279–290 (2009)
Capoor, K., Ambrosi, P.: State and trends of the carbon market. World Bank Institute, http://carbonfinance.org/
Poli, R., Langdon, W.B., McPhee, N.F.: A field guide to genetic programming. Published via http://lulu.com and freely available at http://www.gp-field-guide.org.uk (2008) (With contributions by J. R. Koza)
Silva, S., Costa, E.: Dynamic limits for bloat control in genetic programming and a review of past and current bloat theories. Genet. Program. Evolvable Mach. 10(2), 141–179 (2009)
Lotz, M.: Modelling of process systems with genetic programming. Masters Dissertation, Stellenbosch University (2006), http://etd.sun.ac.za/jspui/handle/10019/570 (downloaded 2008-12-03)
Zhuwakinyu, M.: The Research Unit of Creamer Media (Pty) Ltd., South Africa’s Steel Industry (October 2009), http://www.researchchannel.co.za (2009)
Institute for Global Environmental Studies: EnviroScope (2009), http://enviroscope.iges.or.jp/modules/envirolib/view.php?docid=2136
Carbon Disclosure Project (2009), https://www.cdproject.net/en-US/Results/Pages/overview.aspx
Scaw Metals Group: Sustainable Development Report 2008 (2008), http://www.scaw.co.za/
Silva, S.: A Genetic Programming Toolbox for MATLAB, http://gplab.sourceforge.net/ (2004)
Breiman, L., Friedman, J.H., Olshen, R.A., Stone, C.J.: Classification and regression trees. Wadsworth, Belmont (1984)
Discipulus, http://www.rmltech.com/
MATLAB, http://www.mathworks.com/access/helpdesk/help/toolbox/stats/treeprune.html
Koza, J.R.: Genetic programming – on the programming of computers by means of natural selection. MIT Press, Cambridge (1992)
Luke, S., Panait, L.: Lexicographic parsimony pressure. In: Langdon, W.B., et al. (eds.) Proceedings of GECCO 2002, pp. 829–836. Morgan Kaufmann, San Francisco (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lotz, M., Silva, S. (2010). Application of Genetic Programming Classification in an Industrial Process Resulting in Greenhouse Gas Emission Reductions. In: Di Chio, C., et al. Applications of Evolutionary Computation. EvoApplications 2010. Lecture Notes in Computer Science, vol 6025. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-12242-2_14
Download citation
DOI: https://doi.org/10.1007/978-3-642-12242-2_14
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-12241-5
Online ISBN: 978-3-642-12242-2
eBook Packages: Computer ScienceComputer Science (R0)