IS428 2016-17 Term1 Assign2 Joachim Fu Jun Hao

From Visual Analytics for Business Intelligence
Revision as of 01:08, 26 September 2016 by Joachim.fu.2013 (talk | contribs)
Jump to navigation Jump to search

Abstract

Workplace injuries has always been the concern in all companies as it reflects the effectiveness of its safety practices as well as the welfare of the employees as work. The WSHI National Statistics Report 2014 shows the various types of industries that has a higher proportion of injuries than others due to the nature of the job. In this project, it is important to note many varying factors that contribute to workplace injuries.


Theme of Interest

This project focuses on finding the relationship between the varying factors of work productivity that could lead to the severity and prominence of workplace injuries.


Questions for Investigation

1. How does working overtime affect the productivity which in turn leads to workplace injuries?
2. Does gender determine the suitability and capability of the job role?
3. Does it mean that being experienced means that one has a lower chance of having injuries?

Finding appropriate data attributes

The following data attributes have been selected for analysis:

  • Victims by Gender
  • Cause
  • Body Parts Injured
  • Company
  • Industry
  • Injured When Working Overtime
  • Injured While Performing Official Work Duties
  • Nature of Injury
  • Number of Records
  • Pct Manual Work
  • Occupation
  • Victim's Gender
  • Months worked
  • Victim's Age
  • Number of Records


Data Analysis and Visualisation

1.The first analysis was to determine the victim's age against the number of reported incidents. The purpose is to find out the frequency and distribution of the prominent age groups by viewing the cluster. Another important analysis was on the outliers as it denotes the description of the industry and the body parts injured.

Victim's age Vs Number of Records.JPG


2. The next analysis was to define the different industry types that has different amounts of manual work.

Treemap of Manual Work % and Industry.JPG


3. Experience by months was to determine for the victims' age records, how experienced are they and the severity and prominence of injury.

Experience By Months.JPG


4. Victims by gender against number of records to see the suitability and capability of certain job requirements that the victims' may be be lacking.

Victims By Gender.JPG


5. Overtime injuries against number of records to understand if the workers are doing excessive amounts of work for a particular industry that has already its own strenuous levels on its own.

Overtime Injuries.JPG


6. Body Part Injury against number of records to see what the company could provide in terms of insurance.

Body Part Injury Bar Graph.JPG


7. Performing official work duties against number of records on a bar graph to understand if employees may be given work that is out of their scope and may not have the necessary skills.

Performing Official Work Duties.JPG


8. Cause against number of victims to understand the cause of the injury perhaps due to the low productivity levels of the workers.

Cause Vs Number.JPG


Final Visualisation Dashboard

https://public.tableau.com/profile/publish/JoachimFuAssignment2/Dashboard1#!/publish-confirm

Above is the link for the interactive visualisation of the charts.

Dashboard.JPG

Analysis

Using the table Victim's Age as filter, here are some observations:

  1. The most number of occurences are around the age of 33.
  2. From the outlier of an average age of 33, the most number of occurences at the construction site which has >50% of manual work shows a total number of records at 503 with a prevailing 14.4% who got injured while working overtime, 10% who were injured for doing non-official work and are predominantly male.
  3. Inexperience at work contributes significantly to workplace injury



Tools Utilized

  1. Excel 2013 for data preparation
  2. Tableau for visualization