WIT Press

Derivation Of Safety Measures Through Human Error Risk Analyses


Free (open access)

Paper DOI









340 kb


W. D. Käppler & E. Dalinger


Official statistics state "incorrect behaviour" as cause of most road accidents and conclude that there is hardly anything to do about that. Experts, however, ask why even experienced operators control inadequately or make incorrect decisions. This paper gives an overview over 10 years of related accident and risk analysis research. The underlying accident emergence and risk prediction model treats human error as accident cause. Those errors emerge from causal factors that are already present in any environment and cause latent risks. The analysis of this network leads to the prediction of single causal factor’s risks. Related safety measures are formulated and risk changes predicted on the basis of \“What-If” analyses. At present the analysis covers 777 model items. The methodology comprises action, perception and cognition errors and classifies causal factors into operator disposition and working and environmental conditions. This paper presents model and methodology as well as results. 1 Introduction and problem Statistics show that incorrect behaviour causes 90% of accidents. Publicity is at easy there is nothing you can do about that. Experts, however, know that this may be not the last answer in high technology environments. The question arises why even experienced operators control inadequately, ignore rules and regulations knowingly, oversee important information or make incorrect decisions, since accidents cause: • System loss • Know-how loss • Cost: 25.000 € per accident at work