Prediksi Nilai Cash In PT Telkom Akses dengan Metode Long Short Term Memory Neural Network

Fajrin, Jihan Fitra and Mujiburrahman, Ammar (2024) Prediksi Nilai Cash In PT Telkom Akses dengan Metode Long Short Term Memory Neural Network. Project Report. [s.m], [s.l.]. (Unpublished)

Text 5006211071_5006211079-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

Penelitian ini bertujuan memprediksi arus kas masuk (cash in) PT Telkom Akses (PTTA) menggunakan metode Long Short-Term Memory (LSTM) Neural Network. Data yang digunakan berupa cash in harian periode Januari 2023–Juni 2024, kemudian diolah menjadi data bulanan dan dinormalisasi dengan Min-Max Scaler. Model LSTM dilatih menggunakan 90% data training dan diuji dengan 10% data testing. Evaluasi model menggunakan Mean Absolute Percentage Error (MAPE) dan Root Mean Square Error (RMSE), dengan hasil MAPE sebesar 6,15% dan RMSE 0,15 pada data tereskalasi, menunjukkan akurasi prediksi yang baik. Hasil peramalan 18 bulan ke depan memproyeksikan puncak cash in pada Juli 2024 sebesar Rp 843 miliar, dengan tren penurunan hingga Rp 164 miliar pada Agustus 2025. Temuan ini menegaskan potensi risiko keuangan yang perlu diantisipasi melalui strategi mitigasi, seperti diversifikasi pendapatan dan optimalisasi manajemen arus kas. Keterbatasan penelitian terletak pada periode data yang relatif singkat serta belum mempertimbangkan faktor eksternal yang dapat mempengaruhi arus kas.
============================================================================================================================
This study aims to forecast the cash inflow of PT Telkom Akses (PTTA) using the Long Short-Term Memory (LSTM) Neural Network method. The data used consists of daily cash-in records from January 2023 to June 2024, which were aggregated into monthly data and normalized using the Min-Max Scaler. The LSTM model was trained on 90% of the data and tested on the remaining 10%.Model evaluation was conducted using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE), yielding a MAPE of 6.15% and an RMSE of 0.15 on the scaled data—indicating a good prediction accuracy. The 18-month forecast projects a peak in cash inflow in July 2024, reaching IDR 843 billion, followed by a downward trend to IDR 164 billion in August 2025.These findings highlight potential financial risks that should be mitigated through strategies such as revenue diversification and cash flow management optimization. The main limitation of this study is the relatively short data period and the exclusion of external factors that could influence cash flow.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Cash in, Time Series Forecasting, Neural Network, LSTM, PT Telkom Akses
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Jihan Fitra Fajrin
Date Deposited:	28 Jul 2025 07:56
Last Modified:	28 Jul 2025 07:56
URI:	http://repository.its.ac.id/id/eprint/122117

Actions (login required)

View Item