This data was pulled from their Annual Report tables, which are available on their website. We used a Python script and a tool called pdfTables to extract the data from the annual report into a spreadsheet, which we imported to Power BI.
See a handful relationships and yearly stats at a glance. We can see the number of reports, or warnings, issued by inspectors relative to the numbers of total hours worked by inspectors. We see the number of claims submitted vs claims accepted (won or awarded). We also see claims rejected vs disallowed, disallowed meaning they may need additional information or further review. We also see Injury Rate and Premium Base Rate YoY vs their target.
This shows us the rates of incidences based on Industry Sector and Injury Group (ie. Close Call, Fatal, or Other). We can also see the specific injury type in this list down below. Here we can drill down for more information. For example, looking at Primary Resources, we can see they were quite high leading into 2013 and 2014 but have been trending downwards. If we drill into year, we can see trends in seasonality. Injuries seem to trend upwards through the summer and fall.
These are stack charts showing the top injuries and what activities caused the most injuries. The relative thickness of the colour sections indicates the volume. In the first chart, we can see that close calls are aggressively declining, but other injuries. For example, fatalities are relatively consistent, which should be investigated. Below, we can see that most activities are generally declining based in incidences on the reducing thicknesses of each section, except for House and Wood Framing, which should be investigated.
The past page shows us incidences by Industry and Location so we can see problem sectors and regions that may require more attention or prevention programs.