Business Process Intelligence Berdasarkan Graph Model (Studi Kasus: Enterprise Resource Planning Retail)

Hadiwijaya, Andy Yohanes (2018) Business Process Intelligence Berdasarkan Graph Model (Studi Kasus: Enterprise Resource Planning Retail). Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 5114100050-Undergraduate_Thesis.pdf]
Preview
Text
5114100050-Undergraduate_Thesis.pdf - Accepted Version

Download (1MB) | Preview

Abstract

Event Log merupakan salah satu aset penting di perusahaan. Tugas akhir ini berusaha untuk memodelkan model proses bisnis di event log sehingga bisa menganalisa proses model agar bisa membawa berbagai manfaat bagi perusahaan seperti mengurangi biaya perusahaan, membuat keuntungan meningkat, dan banyak lagi. Salah satu masalah dalam pemodelan event log secara otomatis adalah penambahan non-free choice. Tugas akhir ini bertujuan untuk mengembangkan sebuah sistem yang dapat memodelkan proses bisnis dari event log yang berisi non-free choice dengan menggunakan graph database di Neo4j. Pertama-tama, penelitian ini menghasilkan event log ke dalam link list berupa grafik di Neo4j. Selanjutnya, penelitian ini mengusulkan sebuah metode untuk meningkatkan link list dari event log menjadi graph model yang mengandung non-free choice. Hasil graph model yang diperoleh dengan menggunakan Neo4j dibandingkan dengan hasil yang diperoleh dengan menggunakan LTL yang diturunkan dari aturan dalam model deklaratif. Makalah ini dapat membuktikan bahwa metode yang diusulkan dengan menggunakan graph database adalah metode yang tepat karena hasil grafik memiliki hasil yang lebih baik daripada Heuristic Miner. Hasil Neo4j bisa langsung menghasilkan relasi yang benar meski event log mengandung non-free choice dimana tidak bisa dilakukan pada Heuristic Miner.

========================================================================================================

The event log is one of the important assets in the company. This research strives to model the business process model in event log so that can analyze the model process in order to bring various benefits to companies such as reducing company costs, increasing profits, and more. One of the problem in modeling event log automatically is the addition of non-free choice. This research aims to develop a system that can model a business process of an event log that contains non-free choice using a graph database in Neo4j. First, this research model the event log into a link list in the form of a graph in Neo4j. Next, this research proposes a method to enhance the link list of event log to be a graph model that contains the non-free-choice. The results of the obtained graph model using Neo4j are compared with the results that are obtained using LTL that is derived from rules in a declarative model. This research can prove that the proposed method using graph database is a right method because the graph result has a better result than a Heuristic Miner. The result of Neo4j can directly generate relation which is true even though event log containing non-free choice where can’t be done in the Heuristic Miner.

Item Type: Thesis (Undergraduate)
Additional Information: RSIf 005.276 2 Had b-1 3100018074790
Uncontrolled Keywords: Control Flow Pattern, Non-free-choice, Neo4J, Graph Database
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Electrical Technology > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Andy Yohanes Hadiwijaya
Date Deposited: 27 Mar 2018 03:19
Last Modified: 27 Jul 2020 09:05
URI: http://repository.its.ac.id/id/eprint/50119

Actions (login required)

View Item View Item