Usulan perbaikan kecacatan material kemasan pendukung produk menggunakan pendekatan crisp-dm (cross-industry standard process for data mining)

DETAIL KOLEKSI

Usulan perbaikan kecacatan material kemasan pendukung produk menggunakan pendekatan crisp-dm (cross-industry standard process for data mining)

Oleh : Gina Almas Nabiha

Info Katalog

Penerbit : FTI - Usakti

Kota Terbit : Jakarta

Tahun Terbit : 2024

Pembimbing 1 : Rina Fitriana

Pembimbing 2 : Anik Nur Habyba

Subyek : Electronic packaging--Materials

Kata Kunci : cross-industry standard process for data mining, decision tree, power business intelligence, statist

Saat ini file hanya dapat diakses dari perpustakaan.

Status : Lengkap

Abstrak
Abstract

P Penelitian ini dilakukan karena tingkat kecacatan bahan kemasan pendukung Produk lithos M melebihi standar toleransi Perusahaan yaitu 2%. Penelitian ini bertujuan untuk mengidentifikasi penyebab dan memberikan usulan untuk meningkatkan kualitas bahan kemasan pendukung produk. Metode yang digunakan dalam data mining dengan pendekatan CRISP-DM (Cross-Industry Standard Process For Data Mining).Tahap Business Understanding menentukan masalah dan tujuan penelitian, Power Business Intelligence, SIPOC (Supplier, Input, Process, Output, Customer) Diagram, Operation Process Chart, QC Action, dan CTQ (Critical to Quality). Tahap Data Understanding membuat Peta kendali P, menghitung DPMO dan tingkat sigma yang diperoleh nilai botol penyok mesin unscramble 762.31 dengan tingkat Sigma 4.66, Sticker 2nd defect Internal 187,47 dengan tingkat sigma 5.06, Cap 2nd defect internal 67.18 dengan tingkat sigma 5.32, dan menggunakan Fault Tree Analysis. Tahap Data Preparation melakukan pembersihan data, integrasi, transformasi, dan preprocessing.Tahap Pemodelan melakukan klasifikasi dengan C4.5 dan algoritma Cart decision tree. Tahap evaluasi menggunakan Confusion Matrix dengan akurasi 78.8 persen dan 89.4 persen. Tahap Deployment menghasilkan usulan perbaikan dengan membuat Dashboard, Standard Operating Procedure, dan Check Sheet.Kata Kunci : Cross-Industry Standard Process For Data Mining, Decision Tree, Power Business Intelligence, Statistical Process Control, Fault Tree Analysis.

T This research was conducted because the defect rate of packaging materials supporting lithos M products exceeded the Company\'s tolerance standard of 2%. This research aims to identify the causes and provide suggestions to improve the Quality of product support packaging materials. The methods used in data mining with the CRISP-DM (Cross-Industry Standard Process For Data Mining) approach. The Business Understanding stage determines the problem and research objectives, Power Business Intelligence, SIPOC (Supplier, Input, Process, Output, Customer) Diagrams, Operation Process Chart, QC Action, and CTQ (Critical to Quality). The Data Understanding stage creates a Control P Chart, calculates DPMO and the sigma level obtained by the unscramble machine dented bottle value 762.31 with a Sigma level of 4.66, Sticker 2nd defect Internal 187.47 with a sigma level of 5.06, Cap 2nd defect internal 67.18 with a sigma level of 5.32, and uses Fault Tree Analysis. The Data Preparation stage performs data cleaning, integration, transformation, and preprocessing. The Modelling stage makes classification with C4.5 and the Cart decision tree algorithm. The evaluation stage uses a Confusion Matrix accuracy of 78.8 percent and 89.4 percent, respectively. The Deployment stage produces improvement proposals by creating a Dashboard, Standard Operating Procedure, and Check Sheet.Keywords : Cross-Industry Standard Process For Data Mining, Decision Tree, Power Business Intelligence, Statistical Process Control, Fault Tree Analysis.