Posted By John J There have been a number of discussions regarding target zero and accident rates so how many LTAs per year would you deem acceptable/realistic for a workforce of 1000 in the: manufacturing industry? construction industry? Retail industry?

Posted By John J DP, You have mail John

Posted By jervis Surely no accidents would be acceptable but we live in the real world accidents are usually caused by people!!!!!

Posted By Tom Doyle In a reactive sense, the number of accidents that are acceptable is equal to the number of injuries that you are not willing to prevent. In other words, if after a person is injured at work, and after the accident investigation the aspects contributing to the situation which lead to the injury are understood, the organization does not take real action to prevent a re-occurrence, then one can only conclude that that type of accident is acceptable. Count how many times this happens and you will establish, at least from a reactive perspective, how many accidents are acceptable for your organization. From a proactive perspective, look at your risk assessments. Do they contain enough detail to enable the decision makers to be confident that an accident will not occur? Have all of the preventative measures been verified? If an injury were to occur can the failure mode(s) that contributed to the accident scenario be identified? Were the failure modes taken into account during the risk assessment. Lack of consideration of protective systems failure, engineered and/or behavioural, is a good indicator of how many potential accident scenarios are accepted. In my experience, the more detailed the risk assessment, the more opportunities there are for injury prevention. Cheers Tom Doyle Industrial Safety Integration

Posted By Adrian Watson None!

Posted By Ian Futcher When I took over the Safety role in my organisation in 1999, although we only had one or two riddor per year and thought that wasn't too bad, we thought we ought to put some sort of programme in to do better. However how do you programme for such a low incidence (approx 0.2% of workforce Riddor injured per year)? So we started counting all our first aid injuries (Bird's triangle)and as we work with over 20,000,000 suture needles a year, you can imagine the numbers of small prick-injuries sustained, let alone paper cuts and grazes that caused blood, and had to be treated to prevent contamination of the product. In the first year of this more "forensic" injury reporting and analysis, we injured approx 50% of our people with these small injuries. After much work (assessments, investigations, improvements, training, special awareness activities, safety incentives, hazard spotting forms, behavioural safety activities)we have improved, and in the last 12 months we injured 5% of the employees. Is 5% tolerable? No. Is 0.1% tolerable? maybe to me, but to the injured employees... and for the product security? Probably no. Is 0% achievable? No. Does that mean we shouldn't use 0% as a target? Not at all: 0% is something we should aim for even though it may be impossible to achieve (I know it's not a "SMART" target, but I think it can be an inspirational one). Well that my four-penn'orth Ian

Posted By Andy Brazier Most companies will not have enough serious accidents to actually get any statistically relevant data on which to decide whether their rates are acceptable. For example, the fact that they have not had a fatality or major fire in last 10 years says nothing about whether they will have one tomorrow. It is really only on big projects like the channel tunnel where you can make much use of data. In this case I doubt a prediction was really made. Instead I would expect that historical data regarding similar projects was gathered which said a project of this size, complexity etc. typically results in 24 fatalities. It is a great idea to look at more minor incidents (e.g. first aid) as they do provide more data on which to work on. However, you do need to be very careful because the causes of these minors is often very different to the causes of more serious accidents. This was highlighted in the BP Texas city and Esso Longford accidents. Andrew Hopkins, who has written extensively about the latter makes the point that an airline would not make the mistake of using information about minor injuries to determine how likely it is that their aircraft would crash. To sum up my opinion. The acceptable number of accidents will always be less than the number you had last month/year/decade. But zero accidents does not mean you won't have one tomorrow.