Rancang Bangun Fuzzy Association Rule Miner Untuk Mendeteksi Fraud Pada Proses Bisnis Enterprise Resource Planning (ERP)

Sinaga, Fernandes P. (2014) Rancang Bangun Fuzzy Association Rule Miner Untuk Mendeteksi Fraud Pada Proses Bisnis Enterprise Resource Planning (ERP). Other thesis, Insititut Teknologi Sepuluh Nopember.

Text 5110100190-Undergraduate_Thesis.pdf Download (3MB)

Abstract

Saat ini banyak perusahaan yang telah menggunakan sistem Enterprise Resource Planning (ERP) untuk mengatur proses bisnis yang dijalankan. Proses bisnis ERP pada suatu perusahaan akan selalu mengalami perubahan secara dinamis. Perubahan yang terjadi dapat menghasilkan variasi-variasi terhadap proses bisnis ERP tersebut. Variasi antara proses bisnis yang berjalan terhadap proses bisnis yang standar dapat diperiksa menggunakan process mining. Pada tugas akhir ini, dibahas mengenai variasi proses bisnis yang mengandung kecurangan. Kecurangan pada variasi proses bisnis dapat dideteksi dengan menggunakan metode process mining dan dengan pendekatan fuzzy association rule learning. Process mining mendeteksi kecurangan pada proses bisnis dengan cara memeriksa ketidaksesuaian antara event logs dari proses bisnis berjalan dengan proses bisnis yang sesuai standar perusahaan. Hasil pemeriksaan ketidaksesuaian tersebut berupa kumpulan pelanggaran yang dilakukan terhadap proses bisnis. Kumpulan pelanggaran ini kemudian diolah dengan metode fuzzy association rule learning untuk menghasilkan aturan asosiasi antara perilaku kecurangan yang dilakukan serta mengukur tingkat keparahan yang disebabkan oleh kecurangan yang dilakukan. Akan tetapi, aturan asosiasi yang dihasilkan tidak sepenuhnya dapat mendeteksi kecurangan karena terdapat kemungkinan adanya proses fraud dengan bobot yang rendah dan proses normal dengan bobot yang tinggi. Sehingga dalam penelitian ini ditambahkan aturan tambahan untuk mengatasi masalah tersebut. Dari percobaan yang dilakukan terhadap proses bisnis aplikasi kredit bank dan proses procurement pada ERP, diperoleh akurasi yang tinggi dalam pendeteksian kecurangan yang dilakukan. Hal ini menunjukkan bahwa penggabungan metode process mining dan fuzzy association rule learning dapat digunakan untuk mendeteksi kecurangan pada proses bisnis dengan efektif dan akurat.
=============================================================================================================================
Currently many companies are already using Enterprise Resource Planning (ERP) system to manage running business process. ERP business process in a company will always be changing dynamically. Changes that occur can result in variations to the ERP business process. Variation between business processes, running on standard business process can be checked using process mining. In this final project, we discussed about the variations of business process that contain fraud. Fraud in business process variations can be detected by using process mining and fuzzy association rule learning approach. Process mining can detect fraud on business process by examining the discrepancy between event logs of running business processes with company’s business process standard. The results of compliance checking are collection of offenses commited against business process. This set of violations is then processed by fuzzy association rule learning method to generate association rules between the behaviors of commited fraud and measure the severity caused by the commited fraud. However, the resulting association rules are not fully able to detect fraud because there is a possibility of fraud with the low weight and normal process with a high weight. Thus, in this study we add some association rules to resolve the issue. From the experiments conducted on a bank credit application business process and the ERP procurement process, obtained high accuracy in the detection of fraud commited. This suggest that the combination of process mining method with fuzzy association rule learning can be used to detect fraud in business process effectively and accurately.

Item Type:	Thesis (Other)
Additional Information:	RSIf 005.131 Sin r
Uncontrolled Keywords:	Penggalian Proses, Algoritma Fuzzy Association Rule Mining, Fuzzy Multi Attribute Decision Making, Plugin ProM, Pemeriksaan Kesesuaian, Deteksi Kecurangan, Process mining, Algoritma Fuzzy Association Rule Mining, Fuzzy Multi Attribute Decision Making, Plugin ProM, Conformance Checking, Fraud Detection
Subjects:	Q Science > QA Mathematics > QA76.76.A65 Application software. Enterprise application integration (Computer systems)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	01 Nov 2023 08:48
Last Modified:	01 Nov 2023 08:48
URI:	http://repository.its.ac.id/id/eprint/105038

Actions (login required)

View Item