Safety Metrics

Six Sigma – iSixSigma Forums Old Forums General Safety Metrics

Viewing 4 posts - 1 through 4 (of 4 total)
  • Author
  • #32644


    Can anyone shed some light on some ideas for potential X’s for safety, specific to how you can measure risk behavior?  I’m trying to use DAMIC to reduce injuries, starting with the risky behavior that may be leading to some of these injuries.  My biggest speed bump is trying to find a way to quantify this type of measurement, since it isn’t quite as tangible as collecting cycle times, load rates, etc.  Any help would be greatly appreciated.



    I would look at some of the healthcare industries. I use some of my customers (i.e. hospitals, government agenies, patients, doctors …). When I quantify my metrixs I base my safety on my customers (above). Do you have any sightings? What is the emphasis of the project or the goal/problem of the project? Why do it? Someone started the project because of a problem. Why did they think it was a problem? Was there data that supported their problem?


    Eric BBB

    In the construction industry there is a heirarchy of incidents, depending on the company similar to the following:
    1) Recordable (or near miss)  2) First aid rendered  3) Lost time accident. 
    Usually reasons for the incident are stated in the written reports.  By looking at the data, the reasons or root causes for the various level of incidents you may find a pattern linking certain kinds of unsafe behavior (such as risk taking) to the frequency and/or severity of the incident.


    J. K. JONES

    Picking the “little y” for a safety project seems to be the key.  I generally try number of claims (poisson distribution) or claims incidence rate (incidence rates are usually normally distibuted so you can developa z-score).  Factors to be studied can then be analysed with Prateo Analysis and Chi-squared tables.
    My most sucessful project used the force required to perform a certian task as the “little y.”  This is a continuous variable which allows many more statistical tools to be used.  I used a full factrial designed experiment to analyse this varibale.  

Viewing 4 posts - 1 through 4 (of 4 total)

The forum ‘General’ is closed to new topics and replies.