DOMAIN

Main Menu

UTP Ignite

Project Leader: Prof. Ts. Dr. Jafreezal Jaafar

Co-PI:

AP. Ts. Dr. Izzatdin Abdul Aziz
AP. Ts. Dr. Mohd Hilmi Hasan
AP. Ts. Dr. Norshakirah Ab Aziz
Ts. Dr. Emelia Akashah Patah Akhir

Team Members:

Ade Wahyu Ramadhani
Ku Amirul Ashraf Ku Amir
Hajar Mohd Razip
Fathiah Ruhana Zainonfetry
Nur Syakirah Mohd Jaafar

Department: Computer & Information Science Department (CISD)

Expertise: Big Data Analitics, Predictive Analytics, Machine Learning

DOMAIN

TECHNOLOGY READINESS LEVEL: 8

BRIEF TECHNOLOGY

The DOMAIN solution capable to predict machine failure at least 4 days before, allowing operational engineers to properly plan the repair work by minimising operational downtime.

PROBLEM STATEMENT & CURRENT ISSUES

RM368 million of estimated cost on unplanned shutdown and failure
Challenges equipment causal in identifying potential failures and the failure
Equipment indications can fail without major

USEFULNESS & APPLICATION

The collection of algorithm provide live prediction for 15 critical equipment, such as gas compressors and turbines. It pin-points the sensor that could potentially cause the failure from happening.

The predictive algorithm is trained using 300 billion rows of historical data feeding from 40,000 sensors located at the offshore platforms. Machine Failure prediction accuracy is 4 days ahead.

IMPACT OF THE PRODUCT

The unplanned shutdown of equipment at the oil and gas platforms can be catastrophic in terms of monetary losses. The ability to predict equipment failure before the actual incident happens is the key to cost saving and efficient maintenance.

Through collaboration between UTP-CeRDaS, PETRONAS domain experts experienced seamless learning with UTP-CeRDaS data. PETRONAS Homegrown technology Upskilling talents in data analytics / producing hybrid engineer.

MARKET POTENTIAL

Oil and Gas Industry

INDUSTRY COLLABORATION

Penisular Malaysia Asset (PMA)
PETRONAS CARIGALI

INTELECTUAL PROPERTY (IP)

Copyrights :

LY2021P04932
LY2021P04931

Information For

Admission

Research