Predictive Analytic on Machine Failure by Utilizing Linear Regression (Historical Data)

MOHD SAMLAN, NUR NASHA AYUNI (2020) Predictive Analytic on Machine Failure by Utilizing Linear Regression (Historical Data). [Final Year Project] (Submitted)

[thumbnail of 23318_Nasha Ayuni.pdf] PDF
23318_Nasha Ayuni.pdf
Restricted to Registered users only

Download (3MB)


“Upstream oil and gas is the part of the petroleum industry that locates and
produces crude oil and natural gas through a network of pumps and wells.” Few main
problems in this field are inefficiencies and machine breakdown that occur when wells
are not being optimally engaged, as well as production that stops temporarily when
parts and equipment fail and waiting to be repaired. Machine failure refers to any event
in which it cannot fulfil its mission or task. It may also mean that the machine has
stopped working, is not working properly, or does not meet target expectations. In
addition, the machine mentioned here is referring to turbine generator in oil and gas
industry or PETRONAS itself. Turbine generator is a connection of a shaft of a steam
turbine or gas turbine engine connected to a high-speed electric generator to generate
electricity. Turbine generators are a very important necessity for the oil and gas
industry. These generators provide key energy sources to the industry, in particular to
assist in drilling and digging. Drilling and digging procedures are key to these
industries, and it takes a lot of energy to service heavy equipment. Somehow, the
industry has been experiencing machine failure for years, resulting in interfering with
the smoothness of their production and increase cost of repair after a breakdown.
Hence, the purpose of this study is to predict the machine upcoming failure by utilizing
predictive analytic on machine learning where the process involves linear regression
algorithm, therefore to create a model on KNIME analytic platform. Upon measuring
the accuracy of the predicted model, the result will be displayed as a dashboard for the
user to monitor the condition of the machine state in Power Business Intelligence.

Item Type: Final Year Project
Subjects: Q Science > Q Science (General)
Departments / MOR / COE: Sciences and Information Technology > Computer and Information Sciences
Depositing User: Mr Ahmad Suhairi Mohamed Lazim
Date Deposited: 24 Sep 2021 09:56
Last Modified: 24 Sep 2021 09:56

Actions (login required)

View Item
View Item