The potential of risk assessment based on statistical approach

ABSTRACT: It is typical that safety is a performance area in which it is difficult to plan and control a real improvement. The normal approach to safety improvement is through Risk Assessment. Unfortunately this approach is strongly limited because the evaluation of the risks is based on evaluator’s experience and expertise and so it is affected by low level of reliability. The aim of this paper is to show how the use of statistical problem solving method, such as Six Sigma, could be useful to improve safety level in working activities. In particular, high level of improvement could be obtained by a structured collection of near misses data considering not only the frequency rate of near misses but also the correlation with the root causes or risks. Following several instructions it could be possible to obtain a database of Risk Indicators that could be used for a more reliable Risk Assessment.