Mathematical Geosciences

, Volume 49, Issue 1, pp 121–143 | Cite as

Multivariate Modelling of Geometallurgical Variables by Projection Pursuit

  • E. Sepulveda
  • P. A. Dowd
  • C. Xu
  • E. Addo


The integration of geological and geometallurgical data can significantly improve decision-making and optimize mining production due to a better understanding of the resources and their metallurgical performances. The primary-response rock property framework is an approach to the modelling of geometallurgy in which quantitative and qualitative primary properties are used as proxies of metallurgical responses. Within this framework, primary variables are used to fit regression models to predict metallurgical responses. Whilst primary rock property data are relatively abundant, metallurgical response property data are not, which makes it difficult to establish predictive response relationships. Relationships between primary input variables and geometallurgical responses are, in general, complex, and the response variables are often non-additive which further complicates the prediction process. Consequently, in many cases, the traditional multivariate linear regression models (MLR) of primary-response relationships perform poorly and a better alternative is required for prediction. Projection pursuit is a powerful exploratory statistical modelling technique in which data from a number of variables are projected onto a set of directions that optimize the fit of the model. The purpose of the projection is to reveal underlying relationships. Projection pursuit regression (PPR) fits standard regression models to the projected data vectors. In this paper, PPR is applied to the modelling of geometallurgical response variables. A case study with six geometallurgical variables is used to demonstrate the modelling approach. The results from the proposed PPR models show a significant improvement over those from MLR models. In addition, the models were bootstrapped to generate distributions of feasible scenarios for the response variables. Our results show that PPR is a robust technique for modelling geometallurgical response variables and for assessing the uncertainty associated with these variables.


Geometallurgical modelling Projection pursuit regression Risk management 


  1. Amankwah H, Larsson T, Textorius B (2013) Open-Pit mining with uncertainty: a conditional value-at-risk approach. In: Migdalas A, Sifaleras A, Georgiadis CK et al (eds) Mathematics & statistics. Springer, New York, pp 117–139Google Scholar
  2. Arlot S, Celisse A (2010) A survey of cross-validation procedures for model selection. Stat Surv 4:40–79. doi: 10.1214/09-SS054 CrossRefGoogle Scholar
  3. Armstrong M, Galli A (2011) A new approach to flexible open pit optimisation. Appl Comput Oper Res Miner Ind 24–30Google Scholar
  4. Barnett RM, Manchuk JG, Deutsch CV (2013) Projection pursuit multivariate transform. Math Geosci 46:337–359. doi: 10.1007/s11004-013-9497-7 CrossRefGoogle Scholar
  5. Boisvert JB, Rossi ME, Ehrig K, Deutsch C (2013) Geometallurgical modeling at Olympic dam mine, South Australia. Math Geosci 45:901–925. doi: 10.1007/s11004-013-9462-5 CrossRefGoogle Scholar
  6. Carrasco P, Chiles J-P, Seguret S (2008) Additivity, metallurgical recovery, and grade. In: 8th international geostatistics congress, SantiagoGoogle Scholar
  7. Chatterjee S, Hadi AS (2015) Regression analysis by example. Wiley, New YorkGoogle Scholar
  8. Coward S, Dowd PA (2015) Geometallurgical models for the quantification of uncertainty in mining project value chains. In: 37th APCOM conference, Society of Mining, Metallurgy and Exploration (SME) (ISBN 978-0-87335-417-2), pp 360–369Google Scholar
  9. Coward S, Vann J, Dunham S, Stewart M (2009) The primary-response framework for geometallurgical variables. In: Seventh international mining geology conference, pp 109–113Google Scholar
  10. Coward S, Dowd PA, Vann J (2013) Value chain modelling to evaluate geometallurgical recovery factors. In: 36th APCOM conference, Fundação Luiz Englert, Brazil (ISBN 978-85-61155-02-5), Porto Alegre, pp 288–289Google Scholar
  11. Dowd PA, Xu C, Coward S (2015) Strategic mine planning and design: some challenges and strategies for addressing them. Min Technol 1743286315Y:000. doi: 10.1179/1743286315Y.0000000032 Google Scholar
  12. Dunham S, Vann J (2007) Geometallurgy, geostatistics and project value—does your block model tell you what you need to know? In: Project evaluation conference, Melbourne, pp 189–196Google Scholar
  13. Efron B, Gong G (1983) A leisurely look at the bootstrap, the jackknife, and cross-validation. Am Stat 37:36–48. doi: 10.2307/2685844 Google Scholar
  14. Emery X, Silva D (2009) Conditional co-simulation of continuous and categorical variables for geostatistical applications. Comput Geosci 35:1234–1246. doi: 10.1016/j.cageo.2008.07.005 CrossRefGoogle Scholar
  15. Friedman JH (1984) A variable span smoother. Department of Statistics, Stanford University, StanfordGoogle Scholar
  16. Friedman JH (1994) From statistics to neural networks: theory and pattern recognition applications. In: Cherkassky V, Friedman JH, Wechsler H (eds) Springer, Berlin, pp 1–61Google Scholar
  17. Friedman JH, Stuetzle W (1981) Regression pursuit projection. J Am Stat Assoc 76:817–823CrossRefGoogle Scholar
  18. Friedman JH, Tukey JW (1974) A projection pursuit algorithm for exploratory data analysis. IEEE Trans Comput C-23:881–890Google Scholar
  19. Friedman JH, Stuetzle W, Schroeder A (1984) Projection pursuit density estimation. J Am Stat Assoc 79:599–608CrossRefGoogle Scholar
  20. Ghasemi JB, Zolfonoun E (2013) Simultaneous spectrophotometric determination of trace amount of polycyclic aromatic hydrocarbons in water samples after magnetic solid-phase extraction by using projection pursuit regression. Environ Monit Assess 185:2297–305. doi: 10.1007/s10661-012-2709-7 CrossRefGoogle Scholar
  21. Glover F (1990) Tabu search: a tutorial 1990:74–94Google Scholar
  22. Goulard M, Voltz M (1992) Linear coregionalization model: tools for estimation and choice of cross-variogram matrix. Math Geol 24:269–286. doi: 10.1007/BF00893750 CrossRefGoogle Scholar
  23. Hunt J, Kojovic T, Berry R (2013) Estimating comminution indices from ore mineralogy, chemistry and drill core logging. In: The second AusIMM international geometallurgy conference, Brisbane, pp 173–176Google Scholar
  24. Hunt J, Berry R, Bradshaw D et al (2014) Development of recovery domains: examples from the Prominent Hill IOCG deposit, Australia. Miner Eng 64:7–14. doi: 10.1016/j.mineng.2014.03.014 CrossRefGoogle Scholar
  25. Keeney L, Walters SG (2011) A methodology for geometallurgical mapping and orebody modelling. In: The first AusIMM international geometallurgy conference, Brisbane, pp 217–225Google Scholar
  26. Kumral M (2013) Optimizing ore-waste discrimination and block sequencing through simulated annealing. Appl Soft Comput 13:3737–3744. doi: 10.1016/j.asoc.2013.03.005 CrossRefGoogle Scholar
  27. Lagos G, Espinoza D, Moreno E, Amaya J (2011) Robust planning for an open-pit mining problem under ore-grade uncertainty. Electron Notes Discret Math 37:15–20. doi: 10.1016/j.endm.2011.05.004 CrossRefGoogle Scholar
  28. Lee E, Cook D, Klinke S, Lumley T (2005) Projection pursuit for exploratory supervised classification. J Comput Graph Stat 14:831–846. doi: 10.1198/106186005X77702 CrossRefGoogle Scholar
  29. Lutz S, Hickman S, Davatzes N (2010) Rock mechanical testing and petrologic analysis in support of well stimulation activities at the Desert Peak Geothermal Field, Nevada. In: 35th workshop on geothermal reservoir engineeringGoogle Scholar
  30. Maleki M, Emery X (2014) Joint simulation of grade and rock type in a stratabound copper deposit. Math Geosci 1–25. doi: 10.1007/s11004-014-9556-8
  31. Marcotte D, Caron J (2013) Ultimate open pit stochastic optimization. Comput Geosci 51:238–246. doi: 10.1016/j.cageo.2012.08.008 CrossRefGoogle Scholar
  32. Qianjian G, Jianguo Y (2010) Application of projection pursuit regression to thermal error modeling of a CNC machine tool. Int J Adv Manuf Technol 55:623–629. doi: 10.1007/s00170-010-3114-4 CrossRefGoogle Scholar
  33. Ramazan S, Dimitrakopoulos R (2012) Production scheduling with uncertain supply: a new solution to the open pit mining problem. Optim Eng 14:361–380. doi: 10.1007/s11081-012-9186-2 CrossRefGoogle Scholar
  34. Silverman BW (1984) Spline smoothing: the equivalent variable kernel method. Ann Stat 12:898–916CrossRefGoogle Scholar
  35. Walters SG (2008) An overview of new integrated geometallurgical research. In: Ninth international congress for applied mineralogy, Brisbane, pp 8–10Google Scholar
  36. Weisberg S (2005) Applied linear regression. Wiley, New YorkGoogle Scholar
  37. Xu C, Dowd PA (2009) Conditional simulation of grades controlled by geological indicators. In: Dimitrakopoulos R (eds) Modelling orebody, mine strategic. Planning, Perth, pp 43–49Google Scholar

Copyright information

© International Association for Mathematical Geosciences 2016

Authors and Affiliations

  1. 1.School of Civil, Environmental and Mining EngineeringUniversity of AdelaideAdelaideAustralia
  2. 2.School of Mining EngineeringUniversity of TalcaTalcaChile

Personalised recommendations