Pratama, Fariz Yudha (2022) Perbandingan Peramalan Penjualan Peralatan Kardiologi Menggunakan Metode Machine Learning, Dan Statistika Pada PT.XYZ. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
02411840000127-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (4MB) | Request a copy |
Abstract
Indonesia merupakan negara berkembang dengan jumlah penduduk yang meningkat setiap tahunnya. Hal tersebut membuat permintaan akan fasilitas kesehatan juga ikut meningkat salah satunya yakni alat kesehatan. Alat kesehatan merupakan sesuatu yang tidak dapat dipisahkan dari kehidupan manusia dikarenakan memiliki banyak fungsi. PT.XYZ merupakan salah satu perusahaan yang bergerak dibidang elektromedis berusaha untuk menyediakan alat kesehatan yang dibutuhkan. Semakin meningkatnya jumlah permintaan terhadap alat kesehatan tersebut tidak dapat diimbangi dengan kesiapan perusahaan, Akan tetapi dalam pelaksanaannya terdapat permasalahan dalam melakukan forecasting. Sehingga membuat perusahaan kesulitan untuk mengatasi hal tersebut. Penelitian ini bertujuan untuk melakukan forecasting dengan menggunakan metode machine learning dan statistika dengan 4 timeseries berbeda. Dimana metode machine learning yang digunakan yakni metode LSTM sedangkan metode statistika yang digunakan yakni brown’s double exponential, holtwinter double exponential, dan moving average. Setelah dilakukan pengujian akan dilakukan perbandingkan hasil yang didapatkan dari keempat metode tersebut menggunakan RMSE. Setelah dilakukan penelitian didapatkan hasil RMSE yang terbaik dari masing-masing timeseries yaitu pada timeseries perhari LSTM tensione 16,99 dan ECG300G 1,394. Timeseries 1 bulan yaitu LSTM tensione 63,8806 dan ECG300G 5,752. Timeseries 3 bulan yaitu Brown’s 213,7053 dan MA ECG 13,09568. Dan pada timeseries 6 bulan yaitu Brown’s tensione 194,798 dan MA ECG300G 14,83556.
=====================================================================================================================================
Indonesia is a developing country with a population that increases every year. This makes the demand for health facilities also increase, one of which is medical devices. Medical devices are something that cannot be separated from human life because they have many functions. PT. XYZ is one of the companies engaged in the field of electromedical trying to provide the necessary medical devices. The increasing number of requests for medical devices cannot be matched by the company's readiness. However, in practice there are problems in forecasting. So that makes it difficult for companies to overcome this. This study aims to perform forecasting using machine learning and statistical methods with 4 different time series. Where the machine learning method used is the LSTM method while the statistical method used is brown's double exponential, Holtwinter double exponential, and moving average. After testing, the results obtained from the four methods will be compared using RMSE. After doing the research, the best RMSE results from each time series are the LSTM tensione 16.99 per day and the ECG300G 1.394. The 1-month timeseries are LSTM tensione 63.8806 and ECG300G 5.752. 3-month time series, namely Brown's 213.7053 and MA ECG 13.09568. And at the 6-month time series, namely Brown's tensione 194,798 and MA ECG300G 14,83556.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Alat Kesehatan, Forecasting, Permintaan, RMSE, Timeseries Demand, Forecasting, Medical devices, RMSE, Timeseries |
Subjects: | Q Science Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > QA Mathematics > QA276 Mathematical statistics. Time-series analysis. Failure time data analysis. Survival analysis (Biometry) Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science) |
Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Industrial Engineering > 26201-(S1) Undergraduate Thesis |
Depositing User: | Fariz Yudha Pratama |
Date Deposited: | 10 Feb 2023 02:43 |
Last Modified: | 10 Feb 2023 02:43 |
URI: | http://repository.its.ac.id/id/eprint/96569 |
Actions (login required)
View Item |