PENGEMBANGAN SISTEM IDENTIFIKASI SIMILARITAS PRODUK ENGINEER TO ORDER DENGAN METODE DEKOMPOSISI MATRIKS SVD

• Main Author: Vito Wirawan P.H, Ignatius
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/80833
• Institution: Institut Teknologi Bandung

PT BMT is a company specializing in the production of customized heavy equipment to meet the unique needs of its customers. These needs ranged from assembling to fabricates trailer trucks for coal transportation in mining areas, oil, and gas. The company usually handles thousands of orders each year. The main classification of the work orders are module or spare part fabrication, tractor fabrication, and truck fabrication. However, the varying product demand and its stochastic patterns pose significant challenges for inventory management. Moreover, because the characteristics of engineer to order, the same product could have different material specification in detail, not completely identic. Based on interviews with company personnel, the current process of controlling material and component stock at PT BMT is deemed inefficient. This inefficiency stems from the manual documentation of Material Use Voucher (MUV) documents for data collection on material sharing. To improve inventory management and ensure smooth production and business processes, a more efficient approach is needed, one that leverages modern technology. Researcher have developed a Product Similarity Identification System that relies on material structure parameters and product features or functions, adapted from few research predecessors. The system incorporates the Singular Value Decomposition (SVD) matrix factorization method to identify patterns and characteristics of products systematically and comprehensively. Additionally, codification mechanisms based on product feature parameters and matrix matriculation mechanisms have been developed to identify new products through similarity calculations. The system generates a product network map based on these similarity values, which can be used to document material sharing in inventory management. This approach aims to streamline inventory management processes especially in the material sharing documentation of products and improve the efficiency by 12.5% to 37.5% for existing products and 6.25% to 18.75% for new products variant.