Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/3194
Title: Prediction of Fragment size using Regression Tree and Multi Linear Regression Model
Authors: Pisipati, V K
Bisoyi, S K
Pal, B K
Keywords: Fragmentation
Blasting parameters
Regression tree
Multiple Linear Regression
Issue Date: Dec-2018
Citation: International Conference on Opencast Mining Technology and Sustainability (ICOMS 2018), Singrauli, India, 14-15 December 2018
Abstract: This paper is aimed to develop and compare precise and applicable models based on Regression Tree (RT) and Multiple Linear Regression (MLR) to predict Mean Fragment Size (MFS). In this regard, 35 blasting operations were investigated and the most influential factors on the fragmentation, i.e. Hole diameter, Depth of hole, Spacing, Burden, Stemming length and Specific charge were measured. Also, the Mean Fragment size values for the considered blasting events were carefully measured using WipFrag image analysis software. Regression Tree analysis was done using RapidMiner Studio and Multiple Regression analysis was done using computer-aided solution SPSS (Statistical Package for the Social Sciences) to analyze the data obtained from the study areas. Seven parameters were input into the regression tree and multiple linear regression analysis to generate the model. Mean fragment size (MFS) out of the seven input parameters was dependent variable and the remaining six such as Drill hole diameter (HD), Stemming length (ST), Burden (B), Spacing (S), Specific charge (QC) and Depth of hole (DH) were input as independent variables.The reliability of the developed models was checked using several performance indices, i.e. R2 , MEDAE and RMSE. It is found that the performance indices obtained by the RT model are better compared to the MLR model.
Description: Copyright of this document belongs to proceedings publisher.
URI: http://hdl.handle.net/2080/3194
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2018_ICOMS _BKPal_PredictionFragment.pdfConference paper634.03 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.