Optimasi Multirespon Menggunakan Metode Taguchi Dan Response Surface Methodology Dengan Pendekatan Desirability Function Dan Genetic Algorithm

Rochmanto, Hani Brilianti (2023) Optimasi Multirespon Menggunakan Metode Taguchi Dan Response Surface Methodology Dengan Pendekatan Desirability Function Dan Genetic Algorithm. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6003202010-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (4MB) | Request a copy

Abstract

Jar test adalah suatu percobaan yang berfungsi untuk menentukan dosis optimal dari koagulan yang digunakan pada proses pengolahan air bersih. Proses koagulasi-flokulasi melibatkan zat koagulan dan flokulan dengan kadar yang ditentukan melalui proses jar test. Faktor-faktor yang diteliti di antaranya adalah dosis koagulan, kecepatan pengadukan, dan waktu pengadukan. Kualitas air hasil dari proses jar test dapat diukur melalui pH dan kadar TSS. Dalam mencapai proses jar test yang efektif dan cost-efisien serta menjaga kualitas air yang baik, peneliti bermaksud melakukan optimasi proses jar test dengan desain Taguchi menggunakan pemodelan Response Surface Methodology (RSM) dengan pendekatan Desirability Function (DF) dan Genetic Algorithm (GA). Peneliti bermaksud membandingkan model terbaik yang dihasilkan RSM yang dikombinasikan dengan DF dan GA. Kriteria kebaikan model akan diukur berdasarkan nilai R^2, R_adj^2, Mean Absolute Error (MAE), dan Root Mean Square Error (RMSE) terkecil. Model terbaik akan digunakan untuk menentukan kombinasi level faktor optimal pada proses jar test. Data yang digunakan pada penelitian ini adalah data primer yang diperoleh dari hasil eksperimen proses jar test pada limbah tahu di Laboratorium Teknologi Pengolahan Air, Departemen Teknik Lingkungan, FTSPK ITS pada tanggal 28 November 2022 – 12 Desember 2022. Model ANOVA efek tetap dengan desain Taguchi merupakan model terbaik dengan nilai R^2, R_adj^2, MAE dan RMSE berturut-turut sebesar 99,73%, 98,24%, 0,0167 dan 0,0167. Didapatkan kombinasi level faktor optimal untuk proses jar test adalah Massa Koagulan 5 mg, Kecepatan Pengadukan 200 rpm, Lama Waktu Pengadukan 2 menit, dan Lama Waktu Pengendapan 2,5 menit.
======================================================================================================================================
Jar test is an experiment that serves to determine the optimal dose of coagulant used in the clean water treatment process. The coagulation-flocculation process involves coagulants and flocculants with levels determined through the jar test process. The factors studied include coagulant mass, stirring speed, and stirring time. The water quality resulting from the jar test process can be measured through pH and TSS levels. In order to achieve an effective and cost-efficient jar test process and maintain good water quality, the researcher intends to optimize the jar test process with Taguchi design using Response Surface Methodology (RSM) modeling with Desirability Function (DF) and Genetic Algorithm (GA) approaches. The researcher intends to compare the best model produced by RSM combined with DF and GA. Model goodness criteria will be measured based on the largest R^2, R_adj^2 and the smallest Mean Absolute Error (MAE), and Root Mean Square Error (RMSE) values. The best model will be used to determine the optimal factor level combination in the jar test process. The data used in this study are primary data obtained from the experimental results of the jar test process on tofu waste at the Water Treatment Technology Laboratory, Department of Environmental Engineering, FTSPK ITS on November 28, 2022 - December 12, 2022. The fixed effect ANOVA model with Taguchi design is the best model with R^2, R_adj^2, MAE and RMSE values of 99.73%, 98.24%, 0.0167 and 0.0167, respectively. The optimal factor level combination for the jar test process was obtained as Coagulant Mass 5 mg, Stirring Speed 200 rpm, Stirring Time 2 minutes, and Settling Time 2.5 minutes.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	jar test, orthogonal array, response surface methodology, signal to noise ratio, taguchi
Subjects:	Q Science > QA Mathematics > QA279 Response surfaces (Statistics). Analysis of covariance.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User:	Hani Brilianti Rochmanto
Date Deposited:	17 Feb 2023 01:59
Last Modified:	17 Feb 2023 01:59
URI:	http://repository.its.ac.id/id/eprint/97500

Actions (login required)

View Item