WAHYUDA, OKKY TRI (2026) INTEGRASI ALGORITMA APRIORI DAN DYNAMIC ROP: SISTEM PENDUKUNG KEPUTUSAN INVENTORI PADA TOKO RAOSTEA ALFAZA. S1 thesis, Universitas Mercu Buana Jakarta.
|
Text (HAL COVER)
Cover.pdf Download (396kB) | Preview |
|
![[img]](https://repository.mercubuana.ac.id/style/images/fileicons/text.png) |
Text (BAB I)
BAB 1.pdf Restricted to Registered users only Download (51kB) |
|
|
Text (BAB II)
Bab 2.pdf Restricted to Registered users only Download (213kB) |
|
|
Text (BAB III)
Bab 3.pdf Restricted to Registered users only Download (874kB) |
|
|
Text (BAB IV)
Bab 4.pdf Restricted to Registered users only Download (448kB) |
|
|
Text (BAB V)
Bab 5.pdf Restricted to Registered users only Download (87kB) |
|
|
Text (DAFTAR PUSTAKA)
Daftar Pustaka.pdf Restricted to Registered users only Download (34kB) |
|
|
Text (LAMPIRAN)
Lampiran.pdf Restricted to Registered users only Download (645kB) |
Abstract
Raostea Alfaza Store is a retail grocery business that has accumulated large volumes of sales transaction data yet never optimally utilized it for decisionmaking. Shelf arrangement and stock procurement rely entirely on the owner's intuition, resulting in stock-out and overstock conditions. This study develops a web-based Point of Sale (POS) system integrating the Apriori Algorithm with a multi-layer pruning mechanism and a dynamic per-product Reorder Point (ROP) method as a Decision Support System (DSS) to extract consumer purchasing patterns and automate real-time inventory control. The study employs a quantitative experimental approach using 10,000 unique sales invoices spanning 5 years of active store operations. The system was built using Laravel v12.0 and PostgreSQL v17. The Apriori Algorithm applies a four-layer pruning mechanism: Minimum Support 1%, Minimum Confidence 10%, Lift Ratio > 1 filter, and restriction to frequent 2-itemset. The ROP method applies ROPi = (ADS×LTi) + SSi with Lead Time and Safety Stock configured dynamically per product according to each SKU's supply chain characteristics. The Apriori Algorithm extracted 10 valid association rules, all with Lift Ratio > 1. The strongest rule was found between Buku Gambar A4 SIDU and Pulpen Gel Joyko JK-100 with Lift Ratio 11.71 and Confidence 82.80% - 84.44%. Asymmetric phenomena were identified in the Beng-Beng - Kiko cluster (Confidence 95.86% vs 24.49%) and Biskitop - Kiko (Confidence 94.51% vs 25.88%), identifying both as driver products multiplying Kiko's sales probability by 5.26–5.33 times. The ROP module with dynamic perproduct parameters detected 30 critical-status products (red) out of 160 total monitored products and classified all of them into three color-coded categories: Red (Must Buy Now), Yellow (Prepare to Buy), and Green (Safe Stock). Performance testing recorded average execution times of Apriori 405.91 ms, ROP 253.72 ms, total 659.63 ms against 10,000 transactions. All 10 Black Box Testing sscenarios passed successfully. The system reduced procurement documentation time from 1.5 hours to under 5 minutes. Kata kunci: Apriori Algorithm, Reorder Point, Market Basket Analysis, Decision Support System, Point of Sale, Inventory Control, Multi-layer Pruning. Toko Raostea Alfaza merupakan usaha ritel kelontong yang mengakumulasi data transaksi penjualan dalam jumlah besar namun belum dimanfaatkan secara optimal sebagai dasar pengambilan keputusan. Penataan rak dan pengadaan stok masih mengandalkan intuisi pemilik toko, mengakibatkan stock-out maupun overstock. Penelitian ini membangun sistem Point of Sale (POS) berbasis web yang mengintegrasikan Algoritma Apriori dengan mekanisme pruning berlapis dan metode Reorder Point (ROP) dinamis per produk sebagai Decision Support System (DSS) untuk mengekstraksi pola pembelian konsumen dan mengotomatisasikan pengendalian stok secara real-time. Penelitian menggunakan pendekatan kuantitatif eksperimental dengan dataset 10.000 invoice penjualan unik periode aktif 5 tahun. Sistem dibangun menggunakan Laravel v12.0 dan PostgreSQL v17. Algoritma Apriori dikonfigurasi dengan Minimum Support 1%, Minimum Confidence 10%, filter Lift Ratio > 1, dan pembatasan frequent 2-itemset. Metode ROP menerapkan formula ROPi = (ADS×LTi) + SSi dengan parameter Lead Time dan Safety Stock yang dikonfigurasi dinamis per produk sesuai karakteristik rantai pasok masingmasing SKU. Algoritma Apriori berhasil mengekstraksi 10 aturan asosiasi valid dengan seluruh Lift Ratio > 1. Aturan terkuat pada pasangan Buku Gambar A4 SIDU dan Pulpen Gel Joyko JK-100 dengan Lift Ratio 11,71 dan Confidence 82,80% - 84,44%. Fenomena asimetris ditemukan pada klaster Beng-Beng–Kiko (Confidence 95,86% vs 24,49%) dan Biskitop - Kiko (Confidence 94,51% vs 25,88%), mengidentifikasi keduanya sebagai driver product yang melipatgandakan probabilitas penjualan Kiko 5,26–5,33 kali lipat. Modul ROP dengan parameter dinamis per produk mendeteksi 30 produk berstatus kritis (merah) dari total 160 produk yang dipantau sistem dan mengklasifikasikan seluruhnya ke dalam tiga kategori berbasis warna: Merah (Harus Beli Sekarang), Kuning (Siap-Siap Beli), dan Hijau (Stok Aman). Pengujian performa mencatatkan rata-rata eksekusi Apriori 405,91 ms, ROP 253,72 ms, total 659,63 ms terhadap 10.000 transaksi. Seluruh 10 skenario Black Box Testing dinyatakan berhasil. Sistem berhasil memangkas durasi rekap pengadaan dari 1,5 jam menjadi di bawah 5 menit. Kata kunci: Algoritma Apriori, Reorder Point, Market Basket Analysis, Decision Support System, Point of Sale, Pengendalian Stok, Multi-layer Pruning.
Actions (login required)
 |
View Item |
