Prediksi Kebangkrutan Perusahaan Infrastruktur Berdasarkan Image Financial Ratio Menggunakan Metode Convolutional Neural Networks (CNN)

Nilam, Muhammad Iklil Rafly (2023) Prediksi Kebangkrutan Perusahaan Infrastruktur Berdasarkan Image Financial Ratio Menggunakan Metode Convolutional Neural Networks (CNN). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06311940000079_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (3MB) | Request a copy

Abstract

Saat ini pembangunan infrastruktur terus dilakukan oleh pemerintah Indonesia. Untuk melakukan pembangunan infrastruktur dibutuhkan bantuan dari perusahaan yang bergerak dalam bidang infrastruktur. Terdapat beberapa perusahaan Badan Usaha Milik Negara (BUMN) yang tergolong dalam klaster jasa infrastruktur. Meskipun perusahaan tersebut tergolong dalam perusahaan Badan Usaha Milik Negara (BUMN), perusahaan tersebut masih memiliki risiko untuk bangkrut. Perusahaan dapat dikatakan bangkrut apabila perusahaan tersebut sudah tidak mampu untuk membayar kewajibannya. Dengan melakukan analisis terhadap rasio keuangan suatu perusahaan, dapat diketahui kinerja dari suatu perusahaan. Untuk memprediksi kebangkrutan suatu perusahaan, metode yang sering digunakan adalah Artificial Neural Networks, Support Vector Machine, Altman’s Zscore, dan lainnya. Pada penelitian kali ini akan digunakan metode Convolutional Neural Networks. Data yang digunakan merupakan laporan keuangan masing-masing perusahaan Badan Usaha Milik Negara (BUMN) klaster jasa infrastruktur yang dipublikasikan pada tahun 2019-2021 melalui website masing- masing perusahaan. Hasil yang diharapkan oleh penelitian ini berupa prediksi apakah perusahaan Badan Usaha Milik Negara (BUMN) klaster jasa infrastruktur akan mengalami kebangkrutan atau tidak. Berdasarkan hasil analisis melalui rasio keuangan, perusahaan BUMN klaster infrastruktur memiliki kemiripan pada rasio Total Assets Turnover, Debt to Assets Ratio, dan Gross Profit. Dengan menggunakan metode CNN, didapatkan tingkat akurasi sebesar 71,43% untuk memprediksi suatu perusahaan termasuk kategori akan bangkrut atau tidak bangkrut pada data training. Sedangkan model yang didapatkan oleh data training dapat memprediksi 66,67% dari total data testing.
=================================================================================================================================
Currently, infrastructure development continues to be carried out by the Indonesian government. To carry out infrastructure development, assistance from companies engaged in the infrastructure sector is needed. There are several State-Owned Enterprises (BUMN) companies that belong to the infrastructure services cluster. Even though the company is classified as a State-Owned Enterprise (BUMN), the company still has a risk of going bankrupt. A company can be said to be bankrupt if the company is unable to pay its obligations. By analyzing the financial ratios of a company, it can be seen the performance of a company. To predict the bankruptcy of a company, the methods often used are Artificial Neural Networks, Support Vector Machines, Altman's Zscore, and others. In this study, the Convolutional Neural Networks method will be used. The data used is the financial statements of each State-Owned Enterprise (BUMN) cluster infrastructure services company published in 2019-2021 through the website of each company. The results expected by this study are in the form of predictions whether the cluster infrastructure services State-Owned Enterprises (BUMN) companies will bankruptcy or not. Based on the results of analysis through financial ratios, infrastructure cluster BUMN companies have similarities in the ratio of Total Assets Turnover, Debt to Assets Ratio, and Gross Profit. By using the CNN method, an accuracy rate of 71.43% is obtained to predict a company that will enter the category of going bankrupt or not going bankrupt on the training data. Meanwhile, the model obtained by the training data can predict 66.67% of the total testing data.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Bangkrut, Convolutional Neural Networks, Infrastruktur, Laporan Keuangan, Rasio Keuangan
Subjects:	H Social Sciences > HG Finance H Social Sciences > HJ Public Finance
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis
Depositing User:	Muhammad Iklil Rafly Nilam
Date Deposited:	28 Jul 2023 04:10
Last Modified:	28 Jul 2023 04:10
URI:	http://repository.its.ac.id/id/eprint/100166

Actions (login required)

View Item