IS428 2016-17 Term1 Assign2 Bong Jun Hao

From Visual Analytics for Business Intelligence
Revision as of 03:44, 26 September 2016 by Junhao.bong.2014 (talk | contribs)
Jump to navigation Jump to search

Theme of interest

As of May 2016, the number of workplace deaths has hit 32, 6 more than in the same period last year. This prevalence of workplace-related fatalities has prompted the authorities to raise penalties for offenders or employees with safety lapses, but the question lies in whether stricter or heavier penalties alone can curb this trend.

Furthermore, to what extent would this actually solve the issue? Currently the heavier penalties introduced includes longer stop work time, from 2 weeks increased to 3 weeks, as well as depriving companies of new foreign workers until they have sorted out their safety issues. All these measures only serve to "worsen the dire situation for meeting deadlines and financial penalties for errant companies", according to the Singapore Contractors Association.

Hence, this workplace safety issues has sparked my curiosity in understanding the reason behind these injuries, as well as to find out if there is any underlying relationships between the injuries such that it can be resolved and worked on to effectively reduce the number of such incidents in Singapore, rather than to use financial deterrents which seem to only add on the issues and pressure companies even more.

Questions for investigation

1. When do injuries occur the most throughout the year?
1. Which industry has the highest rate of workplace injuries?
2. Which body part(s) is most commonly injured?
3. Which gender is more prone to injuries?
4. Which age group is more prone to injuries at work?
5. When do injuries occur most often throughout the year?
6. Which time of the day would injuries occur most often?

Data Preparation

The excel file given to us has some irrelevant columns, which I have removed. The Occupier's Name, Org SSIC (2010), 3-digit SSIC (2010) and 2-digit SSIC (2010) has been removed as they are not relevant or useful to the analysis required for this assignment.

Timing of injuries

Distribution of injuries throughout the year

Distribution of injuries across months.png


In 2014, the injuries seem to occur least at the start and the end of the year. January had the lowest number of injuries at 57, followed by February with 246 and then December with 320 injuries.

The distribution of injuries between March to November is relatively fair with slight dips observed from October onwards.

Distribution of injuries by day of week and timing

Distribution of injuries across day by week.png

To better find out and understand if there is a pattern in the injuries that occur across the day, I used a formula to extract out the hours where the injury occurred from the Accident Time column using the formula, "SPLIT(RIGHT(STR([Accident Time]),9),":",1)". This allows me to extract out only the hours where the incident occurred and also to plot a graph using three variables, the day of the week the injury occurred, the hour it occurred as well as how many times it occurred.

From this it can be observed that in general, Sundays tend to have the least number of injuries occurring. An interesting observation is that the peaks of injuries incident seems to occur between 10am-11am on all days, inclusive of Sundays.The second peak where injuries occur seems to be at between 2-4pm.

Profile of injured workers

Age Profile

Distribution of injuries by Age.png

By creating a bin based on the workers' age, setting the bin size to 5 and then converting to continuous, a histogram is obtained to give us a distribution of the age of the injured workers. It can be seen that generally injured workers tend to fall within the age range of 25-30 followed by 30-35 years old.

Experience of workers

Distribution of injuries by experience.png

Experience seems to play a huge factor in determining the proneness of a worker to injury. It seems that majority of injured workers seems to have less than 1 year of working experience. This is especially prominent in the first six months where the number of workers injured with less than 6 months of experience is almost 2.5 times that of workers with at least 1 year of experience. As the workers gain experience as they work, the likelihood of them getting injured seems to fall off.

Injuries by body parts

Distribution of injuries by body parts.png

It seems that the hands are the most susceptible followed by foot injuries. This then leads to the next question as to the nature of injuries that is always linked with these parts, i.e what is the most commonly occurring nature of injuries for the hand?

Injuries by body parts and nature of injury

Distribution of injuries by body parts and nature.png

It can be seen that the most common type and nature of injury would be cuts and bruises followed by crush injuries for the hand. As for the next most common body part injured, the foot, it is most susceptible to crush and then cuts. This gives us an indication of which area of focus we should look at to reduce workplace injuries.

Industry Analysis

Industry with highest number of injuries

It is observed that there are 3 industries in which injuries occur, namely Construction, Manufacturing and the rest categorized as Others. Using a tree map to aid in the visualization shows that Construction related injuries takes up the highest proportion, followed by that in the Others category. For the others category, it can be taken note that jobs in the accommodation and food and services sector as well as that in the wholesale and retail trade sector has higher occurrences of injuries. On the other hand, work related to metal works in the Manufacturing industry is also quite prominent when it comes to injuries.

Occupation most susceptible to injury

Due to the many different kinds of occupation encountering injuries at the workplace, a word cloud is used to quickly visualize the top occupations most susceptible to injuries. The occupations are filtered with a minimum occurrence of 20 so as to give a better and clearer picture. It can be seen that the construction workers are the most prone to injuries, followed by the cleaners and then cooks.