Improving Safety and Reducing Costs

In system design, operation and performance, reductions of cost and of risk are both key. Failures and successes in technological systems are dominated by the human component, as we make mistakes and learn from them, and as we respond to competitive pressures in the marketplace. When we introduce new technology into any homo-technological system, such as high temperature reactors, or any such new products into a market, we take the risk of failure that is dependent on our experience. We can make mistakes and errors; or the product does not have a sufficient competitive advantage. Learning curves are now well known and established in both accident event analysis and product price reduction. We present the facts behind the integral contribution of human error to accidents and risk, and provide the theoretical analysis of outcome rates (undesirable event and cost reduction) based on the Learning Hypothesis. The importance of the accumulated experience on the learning rate, and the depth of experience on the distribution of outcomes, is presented. The implications are given for predicting the reduction of cost and the improvement of safety in homo-technological systems (HTS). In addition, we show how the probability of error can be modeled, including for rare and even unknown events; and we compare and contrast this to the methods presently used for the treatment of human error in probabilistic risk assessment. (author)