Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/220564
Title: DEVELOPING SAFETY LEADING INDICATORS FOR ENERGY AND ENVIRONMENTAL INFRASTRUCTURE USING DATA ANALYTICS
Authors: NEO ZI LIN
Keywords: PFM
Building
Safety
Degree of B.Sc. (Project and Facilities Management)
Project and Facilities Management
Goh Yang Miang
2020/2021 PFM
Work Safety
Data Mining
Leading Indicators
Machine Learning
Energy and Environment Infrastructures
Issue Date: 24-Dec-2020
Citation: NEO ZI LIN (2020-12-24). DEVELOPING SAFETY LEADING INDICATORS FOR ENERGY AND ENVIRONMENTAL INFRASTRUCTURE USING DATA ANALYTICS. ScholarBank@NUS Repository.
Abstract: Work safety has increasingly been an area of focus in recent years, especially in high risk industries. While construction safety has been widely researched and studied, work safety in the manufacturing industry remains largely unexplored. Especially in Energy and Environmental infrastructures such as power, incineration, and desalination plants where workers engage in primary operations that possess high risks. Past research have explored the application of data mining and machine learning techniques in seeking to address work safety issues and improve overall safety performances on sites. Using data analytics, this study aims to identify a suitable machine learning model and useful leading indicators that may be used to predict safety risks and issues in energy and environmental infrastructures. It aims to demonstrate how machine learning techniques may be used for predictive purposes. The leading indicators identified may serve as a guide for organisations to engage in pre-emptive intervention, taking appropriate measures to actively target areas that are susceptible to safety risks. Data for this study was obtained from a local licensed Generation Company registered under the Energy Market Authority (EMA) that oversees the operation of numerous energy and environmental infrastructures. The company adopts a Behaviour Based Safety (BBS) approach, hence this study also explores the integration of BBS data with data analytics to effectively address safety risks on work sites. This study is conducted in accordance with the CRISP-DM (Cross Industry Standard Process for Data Mining) framework to systematically carry out data analysis. Amongst the numerous machine learning models tested, the K-Nearest Neighbour (KNN) classifier was identified as the most suitable machine learning model for the purpose of this study and a combination of types of safety observations was found to be the most useful leading indicator. Since this study was conducted based on data from a single company, further studies will be required to validate its findings. That said, this study serves as a foundation to the application of machine learning techniques and BBS in the work safety realm of the local manufacturing industry.
URI: https://scholarbank.nus.edu.sg/handle/10635/220564
Appears in Collections:Bachelor's Theses

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Neo Zi Lin 2020-2021 .pdf1.99 MBAdobe PDF

RESTRICTED

NoneLog In

Page view(s)

70
checked on Nov 17, 2022

Download(s)

41
checked on Nov 17, 2022

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.