Prediksi Kebutuhan Modal Kerja Menggunakan Metode Time Series Forecasting Pada Perusahaan Bidang Shipping Container Repair

Ulfadi, Alfian Naufal Rizky (2026) Prediksi Kebutuhan Modal Kerja Menggunakan Metode Time Series Forecasting Pada Perusahaan Bidang Shipping Container Repair. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6032241192-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (9MB) | Request a copy

Abstract

Industri pelayaran di Indonesia mengalami pertumbuhan arus kontainer yang pesat, sehingga meningkatkan kebutuhan pemeliharaan untuk menjaga kelancaran operasional pelabuhan. Namun, mayoritas pelaku usaha perbaikan yang merupakan Small and Mid-size Enterprise (SME) dihadapkan pada ketidakpastian biaya operasional dan fluktuasi permintaan layanan. Hal ini menuntut adanya peramalan modal kerja guna memastikan ketersediaan dana optimal dan mencegah krisis likuiditas. Penelitian ini bertujuan meramalkan kebutuhan modal kerja menggunakan pendekatan time series. Metode Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) dan Long Short-Term Memory (LSTM) dievaluasi menggunakan 36 periode data finansial historis untuk menilai tingkat akurasi peramalan tertinggi. Hasil penelitian membuktikan bahwa model Stacked Long Short-Term Memory (LSTM) mengungguli model ARIMAX(0,1,2) dengan nilai Mean Absolute Percentage Error (MAPE) sebesar 11,93% berbanding 18,69% dan Root Mean Square Error (RMSE) sebesar Rp12,73 juta berbanding Rp17,90 juta. Sebagai implikasi manajerial, kemampuan peramalan ini memampukan perusahaan bertransformasi menuju manajemen kas yang proaktif berbasis data. Penelitian ini merekomendasikan adopsi model Stacked LSTM sebagai sistem peringatan dini, pengikatan kontrak harga material jangka panjang, serta penjadwalan tenaga kerja yang fleksibel.
=======================================================================================================================================
Indonesia’s shipping industry has experienced rapid growth in container traffic, increasing the demand for container maintenance and repair services to ensure smooth port operations. However, most container repair businesses primarily Small and Mid-size Enterprise (SME) face uncertainties in operational costs and fluctuations in service demand, making it necessary to forecast working capital needs to maintain optimal liquidity and prevent the risks associated with working capital shortages or surpluses. This study aims to forecast working capital requirements in the container repair industry using time series analysis. Two forecasting methods are employed: Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) and Long Short-Term Memory (LSTM), which are subsequently compared to evaluate the accuracy of their predictions. The study utilizes 36 periods of the company's historical financial data to analyze patterns in working capital needs and to develop a more accurate forecasting model. This approach enables the identification of relevant financial trends and variability as a basis for more effective decision-making. The results demonstrate that the Stacked Long Short-Term Memory (LSTM) model outperforms ARIMAX(0,1,2), achieving a Mean Absolute Percentage Error (MAPE) of 11.93% compared to 18.69% and an RMSE of Rp12.73 million compared to Rp17.90 million, enabling companies to enhance their financial and operational efficiency through data-driven working capital management. Thus, the study aims to contribute to the formulation of more adaptive working capital management strategies within Indonesia’s container repair sector.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Autoregressive Integrated Moving Average with Exogenous Variables, efisiensi operasional, Long Short-Term Memory, modal kerja, perbaikan kontainer, time series, ARIMAX, operational efficiency, LSTM, working capital, container repair, time series
Subjects:	H Social Sciences > HD Industries. Land use. Labor > HD30.27 Business forecasting Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines.
Divisions:	Interdisciplinary School of Management and Technology (SIMT) > 61101-Master of Technology Management (MMT)
Depositing User:	Alfian Naufal Rizky Ulfadi
Date Deposited:	29 Jul 2026 06:30
Last Modified:	29 Jul 2026 06:30
URI:	http://repository.its.ac.id/id/eprint/139557

Actions (login required)

View Item