Author:
Irawan Krisna Risky Putra,Sukmono Tedjo
Abstract
PT. XYZ is engaged in the manufacture and sale of wood veneers. Starting from the constant occurrence of over stock, now the company must make improvements to the production forecasting process so that over stock can be avoided. It can be seen that accurate production forecasting can create conditions for an effective and efficient production system. This study aims to obtain a more accurate forecast of material requirements using the Support Vector Regression (SVR) method, which is the result of the development of a Support Vector Machine (SVM) which has good performance in predicting time series data. Application of the Support Vector Regression (SVR) method with the RBF kernel in predicting the need for veneer production using the MATLAB application, it produces the smallest error rate with a MAPE of 5%, RMSE of 4364.63 and of 0.748274147. on 67 training data and 20 testing data.
Publisher
Universitas Muhammadiyah Sidoarjo
Reference10 articles.
1. S. Wardah and I. Iskandar, “ANALISIS PERAMALAN PENJUALAN PRODUK KERIPIK PISANG KEMASAN BUNGKUS (Studi Kasus : Home Industry Arwana Food Tembilahan),” J@ti Undip J. Tek. Ind., vol. 11, no. 3, p. 135, 2017.
2. D. R. Indah and E. Rahmadani, “Sistem Forecasting Perencanaan Produksi dengan Metode Single Eksponensial Smoothing pada Keripik Singkong Srikandi Di Kota Langsa,” J. Penelit. Ekon. Akutansi, vol. 2, no. 1, pp. 10–18, 2018.
3. A. Adetayo et al., “Jurnal Penelitian Terapan Industri Metode Peramalan untuk Permintaan Penumpang Udara Domestik di Nigeria,” vol. 5, no. 2, pp. 146–155, 2018.
4. A. M. Al’afi, W. Widiart, D. Kurniasari, and M. Usman, “Peramalan Data Time Series Seasonal Menggunakan Metode Analisis Spektral,” J. Siger Mat., vol. 1, no. 1, pp. 10–15, 2020.
5. E. M. Priliani, A. T. Putra, and M. A. Muslim, “Forecasting Inflation Rate Using Support Vector Regression (SVR) Based Weight Attribute Particle Swarm Optimization (WAPSO),” Sci. J. Informatics, vol. 5, no. 2, pp. 118–127, 2018.
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献