Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique
According to the Department of Statistics Malaysia (DOSM) in 2018, manufacturing industry accounted for 91.2% of temporary disability cases and 6.9% of permanent disability cases. Even though there is an increasing number of research on analyzing occupational accidents at automotive manufacturing in...
Saved in:
Main Author: | |
---|---|
Format: | Thesis |
Language: | eng eng |
Published: |
2022
|
Subjects: | |
Online Access: | https://etd.uum.edu.my/10130/1/s819767_01.pdf https://etd.uum.edu.my/10130/2/s819767_02.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my-uum-etd.10130 |
---|---|
record_format |
uketd_dc |
spelling |
my-uum-etd.101302022-12-07T00:32:34Z Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique 2022 Siti Nor Farah Jawahir, Fadzil Mohd Jamil, Jastini Mohd Shaharanee, Izwan Nizal Awang Had Salleh Graduate School of Arts & Sciences Awang Had Salleh Graduate School of Art & Sciences T55-55.3 Industrial Safety. Industrial Accident Prevention According to the Department of Statistics Malaysia (DOSM) in 2018, manufacturing industry accounted for 91.2% of temporary disability cases and 6.9% of permanent disability cases. Even though there is an increasing number of research on analyzing occupational accidents at automotive manufacturing industry in Malaysia, each research aimed for different purposes and methods. This study predicts the tendency of temporary and permanent disability by accurately identifying the characteristics of workplace accidents that occurred within automotive manufacturing in Malaysia. Decision Tree was used to build the predictive modelling of occupational accidents at automotive manufacturing industry. Decision Tree models were constructed with various algorithms (Chi-square, Gini Index and Entropy), numbers of tree branches (two and three) and data partitions (80/20, 70/30 and 60/40). The different models were compared to determine the best model for predicting and identifying the effects of occupational accidents. The best model was a three-branch decision tree model using Chi-Square as the nominal target criterion and 60/40 data partition. The testing accuracy value is 75.52%. The most important variables in the model were types of accident, cause of accidents and job types. This study produced a set of significant factors in explaining safety workplace system and built a predictive model for predicting effect of occupational accidents. It can be served as a guideline to safety management in automotive manufacturing industry in Malaysia. 2022 Thesis https://etd.uum.edu.my/10130/ https://etd.uum.edu.my/10130/1/s819767_01.pdf text eng 2025-03-03 staffonly https://etd.uum.edu.my/10130/2/s819767_02.pdf text eng public other masters Universiti Utara Malaysia |
institution |
Universiti Utara Malaysia |
collection |
UUM ETD |
language |
eng eng |
advisor |
Mohd Jamil, Jastini Mohd Shaharanee, Izwan Nizal |
topic |
T55-55.3 Industrial Safety Industrial Accident Prevention |
spellingShingle |
T55-55.3 Industrial Safety Industrial Accident Prevention Siti Nor Farah Jawahir, Fadzil Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
description |
According to the Department of Statistics Malaysia (DOSM) in 2018, manufacturing industry accounted for 91.2% of temporary disability cases and 6.9% of permanent disability cases. Even though there is an increasing number of research on analyzing occupational accidents at automotive manufacturing industry in Malaysia, each research aimed for different purposes and methods. This study predicts the tendency of temporary and permanent disability by accurately identifying the characteristics of workplace accidents that occurred within automotive manufacturing in Malaysia. Decision Tree was used to build the predictive modelling of occupational accidents at automotive manufacturing industry. Decision Tree models were constructed with various algorithms (Chi-square, Gini Index and Entropy), numbers of tree branches (two and three) and data partitions (80/20, 70/30 and 60/40). The different models were compared to determine the best model for predicting and identifying the effects of occupational accidents. The best model was a three-branch decision tree model using Chi-Square as the nominal target criterion and 60/40 data partition. The testing accuracy value is 75.52%. The most important variables in the model were types of accident, cause of accidents and job types. This study produced a set of significant factors in explaining safety workplace system and built a predictive model for predicting effect of occupational accidents. It can be served as a guideline to safety management in automotive manufacturing industry in Malaysia. |
format |
Thesis |
qualification_name |
other |
qualification_level |
Master's degree |
author |
Siti Nor Farah Jawahir, Fadzil |
author_facet |
Siti Nor Farah Jawahir, Fadzil |
author_sort |
Siti Nor Farah Jawahir, Fadzil |
title |
Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
title_short |
Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
title_full |
Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
title_fullStr |
Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
title_full_unstemmed |
Predicting occupational accident at automotive manufacturing industry in Malaysia using decision tree technique |
title_sort |
predicting occupational accident at automotive manufacturing industry in malaysia using decision tree technique |
granting_institution |
Universiti Utara Malaysia |
granting_department |
Awang Had Salleh Graduate School of Arts & Sciences |
publishDate |
2022 |
url |
https://etd.uum.edu.my/10130/1/s819767_01.pdf https://etd.uum.edu.my/10130/2/s819767_02.pdf |
_version_ |
1776103747930292224 |