Implementasi Time Series Pada Masa Pandemi Covid-19 Terhadap Penjualan Sales Fire Dan Water

Erlangga, Prayoga (2023) Implementasi Time Series Pada Masa Pandemi Covid-19 Terhadap Penjualan Sales Fire Dan Water. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 09211950096003-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (2MB) | Request a copy

Abstract

Pandemi COVID-19 ini telah menuntut perusahaan agar berpikir aktif untuk mengembangkan strategi dan malakukan peramalan dalam menghadapi persaingan bisnis yang semakin ketat. Peningkatan persaingan dan ketidak pastian penjualan juga dialami oleh PT. Indobara Bahana sebagai perusahaan distribusi, rekayasa, dan pengadaan dan kontraktor, pompa, peralatan berputar, deteksi kebakaran & proteksi kebakaran, keselamatan, solusi, dan layanan pengolahan air dan air limbah. Penelitian ini bertujuan untuk mendapat gambaran peramalan Time Series penjualan perusahaan untuk mentukan target proporsional di tahun 2023. Dari bisnis perusahaan akan diteliti dua penjualan sales fire dan water dengan menggunakan metode peramalan ARIMA (Autoregressive Integrated Moving Average) dan ANN (Artificial Neural Network). Penelitian ini menggunakan data kuantitatif dari periode tahun penjualan 2016 hingga Agustus 2022 untuk melihat dan membandingkan peramalan penjualan perusahaan. Metode penelitian untuk meramalkan sales fire dan water, dilakukan menggunakan dua metode, yaitu metode ARIMA dan metode ANN. Metode peramalan pertama menggunakan Metode ARIMA pada sales fire dan water didapatkan model terbaik ARIMA (1,0,1). Hasil dari peramalan terhadap data testing diperoleh nilai MSE sales fire sebesar sebesar 8,763 dan MAPE sebesar 32%, sedangkan sales water diperoleh nilai MSE 106,029 dan MAPE sebesar 50%. Peramalan kedua pada sales fire dan water menggunakan metode ANN didapatkan model arsitektur terbaik keduanya adalah 12-7-1. Hasil dari peramalan terhadap data testing diperoleh nilai MSE sales fire sebesar sebesar 3,950 dan MAPE sebesar 19%, sedangkan sales water diperoleh nilai 56,620 dan MAPE sebesar 25%. Metode ANN yang dipilih untuk meramalkan sales fire dan water dengan diperolehnya nilai akurasi yang terkecil yang menunjukan model yang dihasilkan memiliki hasil permalan yang lebih baik. Penelitian ini diharapkan dapat menjadi salah satu elemen pendukung keputusan terkait tentang penjualan oleh perusahaan
=======================================================================================================================================
The COVID-19 pandemic has forced companies to think actively in developing strategies and forecasting in the face of increasingly fierce business competition. Increased competition and sales uncertainty is also experienced by PT. Indobara Bahana core business is including distributing, engineering, and procurement and contractor company, pumps, rotating equipment, fire detection & fire protection, safety, and Waste water & wastewater treatment solution and services. Reserch purposes to get an overview of the companies sales time series forecasting to determine the proportional target in 2023. From the companies business, two sales of fire and water will be investigated using the ARIMA (Autoregressive Integrated Moving Average) and ANN (Artificial Neural Network) with forecasting methods. This study uses quantitative data from the sales year period 2016 to August 2022 to view and compare the companies sales forecasts. Research Methodology for predicting sales of fire and water is carried out using two methods, ARIMA method and the ANN method. The first forecasting method uses the ARIMA method for sales of fire and water to get the best ARIMA model (1,0,1). The results of forecasting on data testing obtained MSE sales of fire values of 8.763 and MAPE of 32%, while sales of water obtained MSE values of 106.029 and MAPE of 50%. The second forecast for sales of fire and water using the ANN method is that the best architectural model for both is 12-7-1. The results of forecasting on testing data obtained MSE sales of fire values of 3.950 and MAPE of 19%, while sales of water obtained values of 56.620 and MAPE of 25%. The ANN method was chosen to forecast sales of fire and water by obtaining the smallest accuracy value which indicates the resulting model has better forecasting results. This research is expected to be one of the decision support elements related to sales by the company

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Kata kunci: Penjualan, Peramalan, Time Series, ARIMA (Autoregressive Integrated Moving Average), ANN (Artificial Neural Network) Keywords: Sales, Forecasting, Time Series, ARIMA (Autoregressive Integrated Moving Average), ANN (Artificial Neural Network)
Subjects:	H Social Sciences > HA Statistics H Social Sciences > HA Statistics > HA30.3 Time-series analysis
Divisions:	Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT)
Depositing User:	Prayoga Erlangga
Date Deposited:	07 Feb 2023 07:34
Last Modified:	07 Feb 2023 07:34
URI:	http://repository.its.ac.id/id/eprint/96360

Actions (login required)

View Item