Using self-organizing map for quality classification on fish processed product

Muharni, Yusraini (Department of Industrial Engineering)
Hartono, H.M
Ulfah, Maria
Herlina, Lely
Cempakasari, Anita

Journal Industrial Servicess
Vol 9, No 2 (2023): October 2023

Assessing the quality of processed fish products stands as a critical factor in ensuring consumer satisfaction, upholding industry standards, and reducing wastage. Traditional methods for quality classification typically involve manual inspection, which is both time-consuming and subjective. In recent years, the utilization of advanced data analysis techniques, such as Self-Organizing Maps (SOMs), has emerged as a promising approach to enhance the accuracy and efficiency of quality assessment in the fish processing industry. SOMs provide a multi-dimensional map capable of representing various quality attributes of processed fish products. This study aims to classify the quality of processed fish products based on four attributes that impact their time to spoilage. The SOMs effectively segmented the dataset into two clusters, with one cluster being more prone to spoilage, while the other demonstrated a longer shelf life.