Optimization of Jar Test Process Parameters Using Taguchi and Response Surface Methodology with Desirability Function Approach
Hani Brilianti Rochmanto (*a), Muhammad Mashuri (b), Muhammad Ahsan (c)

a,b,c) Department of Statistics, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia
*rochmantohani.206003[at]mhs.its.ac.id

Abstract

The production process of the tofu factory produces wastewater with high levels of organic compound. This compound will reduce water quality if it is discharged directly into the river. This study aims to find the optimal combination of coagulation factor levels so that the waste meets the quality standards before being disposed of or reused. The factors studied included coagulant dosage, agitation speed, agitation time, and settling time. The water result from the jar test process can be measured through these response variables: pH and turbidity levels. Desirability Function (DF) is used to solve multi-response problems, by transforming the two responses into a composite desirability value. The relationship between factors and responses is modelled using second-order Response Surface Methodology (RSM) and Taguchi method, where the goodness of the model is measured based on the largest R2 value, and the smallest Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The best model obtained will be used to determine the optimal factor level combination in the jar test process, by observing the factors which have a significant effect and the contribution of each factor to the model. The results of the analysis showed that the dose of coagulant, agitation time and speed had a significant effect on the changes in turbidity and pH of the waste. Optimum conditions were obtained at the combination of the coagulant dose of 100 mg, agitation time of 2 minutes, and agitation speed of 100 rpm.

Keywords: desirability function, jar test, response surface methodology, taguchi

Topic: MATHEMATICS AND STATISTICS