Decision Tree to Predict the Color Quality of Refined Bleached Deodorized Palm Oil (RBPO)

Abstract

There are many factors that can affect the quality of Refined Bleached Deodorized Palm Oil produced. Every step in RBDPO processing must be mastered very precisely in terms of process conditions such as temperature, flowrate, dose, pressure, etc. This condition can vary from one facility to another and therefore requires the company’s own specifications / rules to implement and integrate all production process facilities until the product is produced. The focus of the study on this particular task is the quality problems faced by companies in the Refinery section where research is carried out using data mining concepts to predict data logsheets that are written every 24 hours by refinery operators. The data can be used as a tool to predict the color quality of Refined Bleached Deodorized Palm Oil (RBPO). The use of Data Mining in this study is to facilitate operators in managing machine settings to achieve the quality required by the company directly. So that the Refinery section does not need to wait for the results of laboratory tests which require approximately 2 hours to find out the RBDPO quality results.