Increasing safety by implementing optimized structures of team communication and the mandatory use of checklists

Abstract

To err is human. This statement is true for all activities where human beings are involved. If safety-critical work has to be performed—independent of the type of industry (aviation, nuclear or medicine)—certain structures and concepts have to be implemented to achieve a high level of safety. Table 1 shows the average mean time between failures (MTBFs) when different tasks have to be performed by human beings. We learn from research work done by Professor Bubb of the Technical University of Munich that we have to expect an average of 15 work errors per person per working day. This is true only for a relaxed working atmosphere. If time pressure and stress increase, the MTBF decreases to one error every half minute. Most of the errors are identified immediately after they happen, but some remain undetected. We need special strategies to neutralize the errors, omissions and mistakes before they cause major problems. A first and important step to reduce the number of errors is to structure the work process. The probability that important work steps are omitted or are done at the wrong time during a certain procedure has to be reduced. The easiest way to help the human operator is to use checklists. They help to identify and correct work errors and omissions before things go wrong or irreversible steps are taken in the wrong direction. As mentioned in Clark et al. [1], checklists have to be tailored to the actual work situation. On the one hand, they have to be as precise and accurate as possible and, on the other hand, they should be as short and self-explanatory as possible. Checklists have to cover safety-critical working steps. Preparing a powerful and helpful checklist is a very difficult and demanding task. Therefore, checklists should be developed only by specialists and professionals who use them in their own work. The use of checklists has to be taught and the procedures have to be updated whenever new information for the system’s optimization becomes available. The use of a checklist has to be emphasized by the top management of an organization and the rules have to be observed by all members of a team. The team leader’s adherence to the procedure is especially important. In aviation, the obligatory use of checklists could increase the safety of a flight by more than a factor of 10. Besides a well-structured working concept, which is ‘protected’ by checklists, there is another area which determines the risk of any operation where human performance is a factor: as shown in Santos et al. [2], we have to deal with communication concepts. A detailed investigation of the work environment combined with the analysis of the flight recorder data and the voice recording of cockpit communication provides a clear picture of the working conditions and the errors that lead to a catastrophe. Accident statistics prove that it is the human being in the cockpit who causes approximately three-quarters of all aviation accidents. The large proportion of human error has given rise to the seemingly brilliant solution of replacing the fallible human being with an ‘infallible’ digitally operated computer. This measure was meant to eliminate all human insufficiencies from the man/machine control loop. A computer never gets tired, is not emotional, does not need a holiday and has a constant level of motivation, etc. A considerable share of human work has been taken over by robots. In many cases, this measure has an increased productivity and has guaranteed an unchanging product quality. In aviation, however, an increased degree of automation has not changed the share of human errors in causing the accidents. Even after the introduction of the so-called HITEC airplanes, the ‘human error’ factor still accounts for 75% of all accidents (Fig. 1). Up to now, the assumption that an increased degree of automation will necessarily lead to an increase in safety has not been proven true. However, if the computer is ruled out as the ultimate safety system, how else can complex operations involving quick and difficult decisions be controlled? We must seek new answers in fields of activity that depend on the smoothest and safest possible interaction of a man and a machine. In this context, the findings of biology, psychology and the social sciences are gaining in importance.