Risk Control Hierarchy

Environmental and Safety Risk Controls

This is the biggest area of concern for me in industry. Most companies are diligent in identifying their risks, but not so for identifying the reduction operational controls.

I ask them all to consider the following. Firstly if they do a semi quantitative analysis of the risk (as per ISO 31000), 1 being low and 5 being high, we then apply this to both probability and then consequence, this will give us a to score of 25 (5×5) for a likely fatality.


Risk Control Hierarchy

Elimination of hazard: examples include the proper disposal of redundant items of equipment that contain substances such as asbestos or PCBs, the removal of excess quantities of chemical accumulated over time in a laboratory, etc. The elimination of hazards is 100% effective. This would equal 25

Suubstitution of hazard: examples include the replacement of solvent-based printing inks with water-based ones, of asbestos insulation or fire-proofing with synthetic fibres or rockwool, the use of titanium dioxide white pigment instead of lead white, etc. The effectiveness of substitution is wholly dependent on the choice of replacement. This would equal 20

Engineering controls: examples include the installation of machine guards on hazardous equipment, the provision of local exhaust ventilation over a process area releasing noxious fumes, fitting a muffler on a noisy exhaust pipe, etc. The effectiveness of engineering solutions is around 70 – 90%. This would equal 15

Administrative controls: include training and education, job rotation to share the load created by demanding tasks, planning, scheduling certain jobs outside normal working hours to reduce general exposure (eg planning demolition and building works during summer recess), early reporting of signs and symptoms, instructions and warnings, etc. The effectiveness of administrative controls ranges from 10 to 50%. They typically require significant resources to be maintained over long periods of time for continuing levels of effectiveness. This would equal 2

Personal protective equipment/informal control: includes safety glasses and goggles, earmuffs and earplugs, hard hats, toe-capped footwear, gloves, respiratory protection, aprons, etc. Their effectiveness in realistic work situations does not exceed 20%. This would equal 1

We now calculate our risk reduction. So for example the risk of electrocution from the use of electrical hand tools is as follows (this is assuming that the risk has been assessed as 25):


  • A RCD is a engineering control = 15                                     Progressive reduction total = 10
  • The use of testing and tagging regularly is administrative =2             Progressive reduction total = 8
  • Training of staff in electrical safety is administrative = 2       Progressive reduction total = 6
  • Routine management inspections is administrative = 2         Progressive reduction total = 4


So now we can identify that the risk is now low, if we just used testing and tagging it is evident that it would still be a high risk.

The trick here is to not bog down in detail, many risk assessments focus too much on likelihood and this is very debatable as to whether it helps us with risk management at all. We all have different opinions on likelihood, but most would agree on consequence of risk. To help here I have added the national safety statistics of how we die at work (WORK-RELATED TRAUMATIC INJURY FATALITIES, AUSTRALIA 2006–07, Commonwealth of Australia 2009, ISBN 978-0-642-32926-4)

This may help in the debate, because on site it is clear that working at heights and being in a vehicle accident are the most likely way to die at work. A similar approach can be taken for environmental issues, the biggest issue probably being pollution of water and soil (no data found).

Author credit to David Lee, Original post here.