SISTEM PAKAR BERBASIS FORWARD CHAINING DAN CERTAINTY FACTOR UNTUK REKOMENDASI TINDAKAN MAINTENANCE MESIN KRAFT DI PT DIRGANTARA INDONESIA

Jenni Febiyola Sari, - and Liptia Venica, - and Dewi Indriati Hadi Putri, - (2025) SISTEM PAKAR BERBASIS FORWARD CHAINING DAN CERTAINTY FACTOR UNTUK REKOMENDASI TINDAKAN MAINTENANCE MESIN KRAFT DI PT DIRGANTARA INDONESIA. S1 thesis, Universitas Pendidikan Indonesia.

Abstract

Proses maintenance merupakan aspek fundamental dalam menjaga keberlangsungan operasional dan efisiensi produksi di sektor industri, khususnya dalam manufaktur kedirgantaraan. Salah satu mesin yang memiliki peran strategis adalah mesin Kraft, yang digunakan secara intensif di PT Dirgantara Indonesia untuk produksi komponen pesawat. Namun, kondisi mesin yang telah menua serta keterbatasan teknisi berpengalaman menimbulkan tantangan signifikan dalam pelaksanaan maintenance. Studi ini mengusulkan pengembangan sistem pakar berbasis web yang mengintegrasikan metode inferensi Forward Chaining dengan pendekatan Certainty Factor (CF) untuk membantu teknisi pemula dalam pengambilan keputusan tindakan maintenance secara akurat. Metode Forward Chaining digunakan untuk menelusuri aturan secara progresif berdasarkan input kondisi mesin, sedangkan Certainty Factor diterapkan untuk merepresentasikan tingkat keyakinan pakar terhadap suatu diagnosis, sehingga sistem mampu menangani ketidakpastian yang muncul dalam proses penalaran. Sistem dirancang untuk memproses input seperti subsystem, device, component, failure type, dan root cause, lalu menghasilkan rekomendasi tindakan maintenance. Hasil yang diharapkan adalah terciptanya alat bantu pengambilan keputusan yang meningkatkan presisi maintenance, mengurangi ketergantungan pada pakar yang telah pensiun, serta memperkuat keandalan jangka panjang operasi mesin Kraft. ----- Maintenance processes are fundamental to ensuring operational continuity and production efficiency across industrial sectors, particularly in aerospace manufacturing. One of the machines with a strategic role is the Kraft machine, which is extensively used at PT Dirgantara Indonesia for aircraft component production. However, its aging condition and the limited availability of experienced technicians pose significant challenges to effective maintenance execution. This study proposes the development of a web-based expert system that integrates the Forward Chaining inference method with the Certainty Factor (CF) approach to assist novice technicians in making accurate maintenance decisions. Forward Chaining is employed to trace rules progressively based on machine condition inputs, while Certainty Factor is applied to represent the expert’s degree of belief in a given diagnosis, enabling the system to handle uncertainty in the reasoning process. The system is designed to process inputs such as subsystem, device, component, failure type, and root cause, and to generate targeted maintenance recommendations. The expected outcome is a decision-support tool that enhances maintenance precision, reduces reliance on retired experts, and strengthens the long-term reliability of Kraft machine operations.

	Text S_MKB_2100175_Title.pdf Download (1MB)
	Text S_MKB_2100175_Chapter1.pdf Download (528kB)
	Text S_MKB_2100175_Chapter2.pdf Restricted to Staf Perpustakaan Download (1MB)
	Text S_MKB_2100175_Chapter3.pdf Download (2MB)
	Text S_MKB_2100175_Chapter4.pdf Restricted to Staf Perpustakaan Download (1MB)
	Text S_MKB_2100175_Chapter5.pdf Download (550kB)
	Text S_MKB_2100175_Appendix.pdf Restricted to Staf Perpustakaan Download (9MB)

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?hl=en&user=gKOKUgoAAAAJ ID SINTA DOSEN Pembimbing: Liptia Venica: 6779029 Dewi Indriati Hadi Putri: 6720737
Uncontrolled Keywords:	Sistem Pakar, Certainty Factor, Forward Chaining, Maintenance Expert System, Forward Chaining, Certainty Factor, Maintenance
Subjects:	T Technology > T Technology (General)
Divisions:	UPI Kampus Purwakarta > S1 Mekatronika dan Kecerdasan Buatan
Depositing User:	Jenni Febiyola Sari
Date Deposited:	01 Dec 2025 03:45
Last Modified:	01 Dec 2025 03:45
URI:	http://repository.upi.edu/id/eprint/145557

Actions (login required)

View Item