Yola, Muhammad Meikel (2026) Implementasi Regresi Linear Untuk Memprediksi Penjualan Produk Pada Toko Nursila Bakery. S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.
![[img]](http://repository.wicida.ac.id/style/images/fileicons/text.png) |
Text
2043125-S1-Jurnal.pdf Download (1MB) |
|
Text
2043125-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Ketidakpastian permintaan konsumen merupakan tantangan utama bagi Usaha Mikro, Kecil, dan Menengah (UMKM), termasuk Toko Nursila Bakery yang bergerak di bidang industri kuliner. Selama ini, perencanaan produksi di toko tersebut masih mengandalkan intuisi pemilik, yang berisiko menyebabkan kelebihan stok atau hilangnya peluang keuntungan. Penelitian ini bertujuan untuk membangun sistem prediksi penjualan kue kering berbasis web menggunakan algoritma Regresi Linear dengan metodologi CRISP-DM (Cross-Industry Standard Process for Data Mining). Data yang digunakan berasal dari pencatatan penjualan historis periode Idul Fitri tahun 2021-2025. Data Idul Adha dieliminasi pada tahap persiapan data karena hanya tersedia untuk rentang waktu 4 tahun, sementara kriteria minimal analisis time series dalam penelitian ini adalah 5 tahun. Analisis dilakukan melalui dua pendekatan: Regresi Linear Sederhana untuk produk tunggal dan Regresi Linear Berganda dengan variabel interaksi untuk membandingkan produk berdasarkan kategori ukuran. Hasil analisis data menggunakan SPSS terhadap berbagai varian produk menunjukkan bahwa seluruh model regresi signifikan secara statistik dengan nilai Sig. < 0,05. Model ini menghasilkan koefisien determinasi (R2) yang kuat, berkisar antara 0,863 pada produk Nastar Sedang hingga angka tertinggi 0,990 pada produk Nastar Kecil. Pada model berganda, kombinasi Semprit Susu dan Putri Salju Vanilla (Ukuran Sedang) memiliki akurasi tertinggi dengan nilai R2 sebesar 0,936. Implementasi sistem berbasis web ini diharapkan dapat membantu manajemen dalam merencanakan strategi produksi dan persediaan secara lebih akurat serta memfasilitasi pencatatan transaksi toko melalui fitur pemasukan dan pengeluaran. ============================================================ Uncertainty in consumer demand is a major challenge for Micro, Small, and Medium Enterprises (MSMEs), including Nursila Bakery, which operates in the culinary industry. To date, production planning at this store has relied heavily on the owner’s intuition, which poses risks such as overstocking or missed profit opportunities. This study aims to develop a web-based cookie sales prediction system using the Linear Regression algorithm and the CRISP-DM (Cross-Industry Standard Process for Data Mining) methodology. The data used in this research are derived from historical sales records during the Idul Fitri period from 2021 to 2025. Sales data from the Idul Adha period were excluded during the data preparation stage because they were only available for a four-year span, whereas the minimum requirement for time series analysis in this study is five years. The analysis was conducted using two approaches: Simple Linear Regression for single-product prediction and Multiple Linear Regression with interaction variables to compare products based on size categories. The results of data analysis using SPSS across various product variants indicate that all regression models are statistically significant, with Sig. values below 0.05. The models produced strong coefficients of determination (R²), ranging from 0.863 for the medium-sized Nastar product to a maximum of 0.990 for the small-sized Nastar product. In the multiple regression model, the combination of Semprit Susu and Putri Salju Vanilla (medium size) achieved the highest accuracy, with an R² value of 0.936. The implementation of this web-based system is expected to assist management in planning production and inventory strategies more accurately, while also facilitating store transaction recording through income and expense management features.
| Item Type: | Thesis (S1 Teknik Informatika) |
|---|---|
| Additional Information: | Pembimbing 1 : Hanifah Ekawati, S.Pd.,M.Pd Pembimbing 2 : Ahmad Abul Khair, S.Kom.,MT |
| Uncontrolled Keywords: | Regresi Linear, Prediksi Penjualan, CRISP-DM, UMKM |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
| Divisions: | Teknik Informatika |
| Depositing User: | Mr. Muhammad Meikel Yola |
| Date Deposited: | 30 Jan 2026 01:50 |
| Last Modified: | 30 Jan 2026 01:50 |
| URI: | http://repository.wicida.ac.id/id/eprint/6284 |
Actions (login required)
 |
View Item |
