Analyzing non-stationarity in cement stone pit by median polish interpolation: a case study

Tutmez B.

JOURNAL OF APPLIED STATISTICS, cilt.41, ss.454-466, 2014 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 41 Konu: 2
  • Basım Tarihi: 2014
  • Doi Numarası: 10.1080/02664763.2013.840274
  • Sayfa Sayıları: ss.454-466


The raw materials utilized in the manufacture of cement comprise mainly of lime, silica, alumina and iron oxide. Spatial evaluation of these main chemical constituents of cement has crucial importance for providing effective production. Because these components are composed of some raw materials such as limestone and marl, the spatial relationships in a calcareous marl stone pit was taken into consideration. In practice, spatial field data taken from a cement quarry may include some variations and trends. For modeling and removing spatial trend in a cement raw material quarry as well as providing unbiased estimates, median polish kriging was used. By using the variation of the data itself, some approximations and interpolations were carried out. It was recorded that the method obtained outlier-resistant estimation of spatial trend without needing an external exploratory variable. In addition, it provided very effective estimations and additional information for analyzing spatial non-stationary data.