Forecasting With Recurrent Neural Networks For Intermittent Demand Data

Amri, Muhaimin (2021) Forecasting With Recurrent Neural Networks For Intermittent Demand Data. Masters thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
06211850012002_Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

Data permintaan berselang biasanya disebut data permintaan pelanggan atau data penjualan. Dataset mencatat nilai bukan nol jika ada permintaan. Jika tidak ada permintaan, catatan set data adalah nilai nol. Masalah umumnya adalah bahwa permintaan tidak selalu terus-menerus tetapi terputus-putus. Karakteristik ini membuat data yang terputus-putus sulit digunakan untuk prediksi. Metode standar yang digunakan untuk memprediksi data permintaan berselang antara lain Croston dan single exponential smoothing (SES). Croston dan SES biasanya menghasilkan prakiraan statis. Penelitian ini memanfaatkan metode deep learning Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU), dan Long Short Term Memory (LSTM) untuk memprediksi data intermiten. Studi simulasi dilakukan dengan menghasilkan dataset dengan 6 parameter desain yang berbeda dan dengan 50 pengulangan. Selain itu studi empiris menggunakan data dari website Kaggle. Penelitian ini mengukur kinerja prediksi data permintaan intermiten dengan RNN, GRU, dan LSTM, dibandingkan dengan Croston dan SES sebagai metode benchmark. Pengukuran kinerja evaluasi menggunakan mean absolute error (MAE) dan root mean squared scaled error (RMSSE). Dalam studi simulasi, sebagian besar metode jaringan saraf berulang dapat bekerja dengan baik dalam skor MAE. Untuk studi empiris, metode jaringan saraf berulang mengungguli metode konvensional dalam skor MAE untuk semua kumpulan data. Namun, metode konvensi Croston bekerja dalam skor RMSSE untuk sebagian besar studi simulasi dan satu studi empiris. ========================================================================================================= Intermittent demand data is usually called customer demand data or sales data. The dataset will record a nonzero value if there is a demand. If there is no demand, the dataset records are zero values. The general problem is that demand is not always continuous but intermittent. This characteristic makes intermittent data difficult to use for prediction. Standard methods used to predict intermittent demand data include Croston, single exponential smoothing (SES), and others. The Croston and SES typically produce static forecasts. This study utilized deep learning methods recurrent neural network (RNN), gated recurrent units (GRU), and long short-term memory (LSTM) to predict intermittent data. The simulation study was carried out by generating datasets with 6 different design parameters and with 50 repetitions. Besides, the empirical study used data from the Kaggle website. This study measured the performance of predicting intermittent demand data by RNN, GRU, and LSTM, comparison to Croston and SES as the benchmark methods. The performance measurements included the evaluation of mean absolute error (MAE) and root mean squared scaled error (RMSSE). In simulation studies, most recurrent neural network methods can perform well in MAE scores. For the empirical study, recurrent neural network methods outperform conventional methods in MAE scores for all datasets. Yet, the convention method of Croston works in RMSSE scores for most simulation studies and one empirical study.

Item Type: Thesis (Masters)
Uncontrolled Keywords: peramalan permintaan, permintaan terputus-putus, neural network, recurrent neural network, time-series demand forecasting, intermittent demand, neural network, recurrent neural network, time-series
Subjects: Q Science
Q Science > QA Mathematics
Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry)
Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49101-(S2) Master Thesis
Depositing User: Amri Muhaimin
Date Deposited: 10 Mar 2021 01:41
Last Modified: 10 Mar 2021 01:41
URI: https://repository.its.ac.id/id/eprint/83999

Actions (login required)

View Item View Item