Implementasi Fuzzy Logic untuk Menentukan Weight Compensation pada Sistem Auto Weighing Screw Feeder

Hidayat, Febrian Robbani (2026) Implementasi Fuzzy Logic untuk Menentukan Weight Compensation pada Sistem Auto Weighing Screw Feeder. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040221064_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Penelitian ini bertujuan mengembangkan sistem Auto Weighing untuk proses penimbangan bahan baku (raw material) kampas rem di PT Indoprima Gemilang Plant 2 guna membantu mengurangi potensi overweight yang mengakibatkan reject pada proses penimbangan sebesar 0,05% pada Q1 2026. Sistem dirancang menggunakan PLC Mitsubishi FX3U sebagai pengendali utama screw feeder dengan dua mode pengisian, yaitu bulk feed (kecepatan tinggi sebesar 50Hz) dan dribble feed (kecepatan rendah sebesar 20Hz), sehingga aliran material dapat dikendalikan secara lebih presisi. Penentuan titik henti motor dilakukan menggunakan metode Fuzzy Inference System (FIS) Mamdani berdasarkan dua parameter masukan, yaitu Error Prev dan Flow Rate, untuk menghasilkan nilai Weight Compensation (WKomp) sebagai acuan penghentian motor sebelum target penimbangan 16.500 gram tercapai. Pengujian dilakukan untuk mengevaluasi kinerja sistem Auto Weighing berbasis Fuzzy Inference System (FIS) Mamdani berdasarkan parameter error penimbangan, ketepatan pencapaian target berat 16.500 gram, dan waktu siklus penimbangan. Hasil pengujian menunjukkan bahwa sistem menghasilkan nilai Mean Absolute Error (MAE) sebesar 5,4 gram pada material pasir dan 21,1 gram pada material tepung, dengan hasil penimbangan berada pada rentang 16.450-16.550 gram. Waktu siklus yang diperoleh berkisar antara 47-50 detik pada kondisi aliran material normal dan 53-62 detik pada kondisi material menggumpal. Implementasi metode ini juga mampu mengurangi kejadian overweight dan underweight, serta meningkatkan ketepatan penentuan titik henti motor sebesar 84% pada target setpoint 16.500 gram, sehingga proses penimbangan menjadi lebih stabil terhadap perubahan karakteristik aliran material.
======================================================================================================================================
This study aims to develop an Auto Weighing system for the weighing process of raw materials for brake pads at PT Indoprima Gemilang Plant 2 to help reduce the potential for overweight conditions that result in rejects during the weighing process by 0.05% in Q1 2026. The system is designed using a Mitsubishi FX3U PLC as the main controller for the screw feeder with two filling modes: bulk feed (high speed at 50 Hz) and dribble feed (low speed at 20 Hz), allowing for more precise control of material flow. The motor stop point is determined using the Mamdani Fuzzy Inference System (FIS) based on two input parameters Error Prev and Flow Rate to generate a Weight Compensation (WKomp) value as a reference for stopping the motor before the target weight of 16,500 grams is reached. Tests were conducted to evaluate the performance of the Mamdani Fuzzy Inference System (FIS)-based Auto Weighing system based on weighing error parameters, the accuracy of reaching the 16,500-gram target weight, and the weighing cycle time. The test results showed that the system produced a Mean Absolute Error (MAE) of 5.4 grams for sand and 21.1 grams for flour, with weighing results falling within the range of 16,450-16,550 grams. The cycle times ranged from 47-50 seconds under normal material flow conditions and 53-62 seconds when the material was clumped. The implementation of this method also reduced instances of overweight and underweight conditions and improved the accuracy of the motor stop point determination by 84% at the 16,500 gram target setpoint, making the weighing process more stable in the face of changes in material flow characteristics.

Item Type:	Thesis (Other)
Subjects:	T Technology > TS Manufactures > TS176 Manufacturing engineering. Process engineering (Including manufacturing planning, production planning) T Technology > TS Manufactures > TS183 Manufacturing processes. Lean manufacturing.
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Febrian Robbani Hidayat
Date Deposited:	05 Aug 2026 04:37
Last Modified:	05 Aug 2026 04:37
URI:	http://repository.its.ac.id/id/eprint/144025

Actions (login required)

View Item