A very warm welcome to our current topic, Workload and Human Factors-Analysis. I would like to present to you topics around the man and the human factors topic. My name is Harald Schaub. Workload is the amount of work an individual has to do. Actual amount of work and individual perception is what workload makes. Assessment of operator's workload has a retail impact on the design of the human-machine system. Occupational or work-related stress is the response people may have when they're presented with work demands and pressures that are not matched to the knowledge or abilities, and which challenge their abilities to cope with. This may lead to inefficiency, to inadequate man-machine co-operations, and to human errors. Prevention of inadequate workload and occupational stress, increases efficiency, reliability, and safety of complex systems and support accomplishment of mission objectives. Human errors, 70 percent of industrial accidents are directly are caused by human errors, 70 percent. This is due to bad design of the technical system. For example, the system is non-intuitive. This is due to inadequate or failed processes. For example, not supporting situation awareness. This is due to management problems. For example, unclear lines of responsibility. This is due to improper training. For example, not training of problem-solving skills. Thirty percent of industrial accidents are caused by failures in technology in organization. But let us puts a question, who designs the system? Who manages the system? Who maintains the a technology. Does this also people who runs a technology, who builds a technology and runs the organization? Human error is not intended by the actor, it's desired by a set of rules or external observers. It leads a task or a system outside its acceptable limits. Human errors are not only the primary cause of contributing factors in disasters and accidents in industry, but also the major contribution in inefficiency. Human error is a deviation from intention, from expectation, or from desirability. We are able to evaluate workload. Evaluating the workload as the operators is doing the design of the new system, we can iterating with existing systems, and with problems such as workload bottlenecks, overload, and inefficiency can be identified. The human error is a central part of the human-machine system. The correction of these problems is necessary for the operations for a safe and efficient system. Mental and physical stress, inefficiency and unsafe acts. The Workload or Human Factors-Analysis identified the human issues for mental and physical stress, for inefficiency and unsafe acts. We can do this on several levels, we can do this on a skill-based levels where action, communication, man-machine-interactions occur in the operator's execution of a certain routine, or on a decision level where actions, communications, where man-machine-interactions occur when the operator proceed an intended set of actions, an intended set of measures. Or on a perceptual level, where actions, communications, and interactions may occur when the operator's sensory input is degraded or when decisions made upon faulty information. We have certain preconditions for stress, for inefficiency, for unsafe. We have preconditions which are related to the environment, for the physical environment, the technical environment. We have preconditions which are related to the condition of the operator himself, of his current mental state, of his physiological state, of his mental limitation. We have preconditions which are related to certain other personal factors. For example, crew or team factors, or readiness level of people or training level of people. In a workload and stress situations, there are certain aspects we can control when we designed the system. The first part we can control is, is the system suitable for the task? Is the complexity of this operation, is the functional range, is the accomplishment requirements, is it really built up within the system? Second aspect we can control to monitor and to design a situation is the question, is the system self description? Has it a self-description ability? Has the chance that the operator understand what the system says? What the system do? The third aspect is controllability. Which functions are controlled by whom? By the operator? By the system? That imagine a car has the steering wheel, controllability is to the person. If you have another function within a car, for example, the inner state of the engine, this is controlled by technology. The first aspect is expectation conformation. Is a system behave like the operator expect? Is a cognitive expectation of the operator fit or matched to what the system really do or what the system may do in the future. Error robustness, is a system tolerant in relation to errors done by the operator? Does it correct the errors? Does it monitor the errors? Does it give it a chance to rehearse the errors? Individual barness, is a system adaptability to the owner or to the operator's desire with colors, with font size, but also with the overall behavior. Is the system's possibility to learn? To learn how to control the system? To learn how to control the functions done by the system, especially the last past? Learning from the system may come crucial in our new system which are more or more running in an automated way where the operator has nothing to do, and so he has no chance to learn from the system because most situation the system is doing all the function by itself. A crucial point is the rule of the organization at all and the special offices supervision and of the management in stressful, inefficient, or in unsafe situation. We quite often have inadequate supervision where the management will not take it as a role or the task they have to do in controlling and managing situation. We have quite often badly or inappropriate planned operations where the plan is not really fitting to the requirements of the situation, and sometimes you can have the impression that the plan is designed to fail because it's inappropriate. We have failing correcting known problems where people will not start problem-solving, just ignoring the problem and the problem becomes bigger and bigger and bigger. We have supervision in violations where the management will not act as they have to do, will not act in controlling this situation, will not act in giving guidance, will not act giving goals and aims. The factors controlling all of this is related to resource management, which is related to the management of the technical resources, but also of course, related to the management of the personal resources, of human resources. The crucial point is fitting these things together. The technical point and the personal point or the human factors point, fit together. That the right technology for the right tasks, with the right people are on the right time in charge. Perhaps, one of the most crucial aspects is the organizational climate, the organizational culture. Is there an error culture of talking about errors, of reporting errors, but not blamed to make errors. Is there a culture of talking about their own experience of stress, of inefficiency or is there a culture of not talking about these things or ignoring these things and dealing with these things when it happens or not in advanced? You have to put all these things in an operational process, where you can expect that people act different when they are under stress, they act different when they see or when they expect to be inefficient, that they act different when they see or when they produce unsafe acts. You can expect this or you can manage this with appropriate operational process. We have certain methods to analyze human factors at all our special workload. One of the most appropriate method is direct observation within the actual situation on the workplace, in the situation, on a board of a submarine or within a ship, or in a plane, or within a car, or in a plant or in a team situation. What we are doing there is observing the people with psychological and with subject matter experts observers. We use technical systems to directly observe. One very appropriate method is eye tracking, where you can focus where people's conscious is, where the amount of workload is and you have certain physiological measures. You can do also heart rate for example, and of course, you can do interviews talking to the people, observing what they are doing. It is very accurate when you are within the situation, you have a very valid and good access to the workload because you measured within the situation itself. The problem is, it is quite costly and it's consuming time in the realization, the system needs to be there. For example, if you are just building a new system, you cannot do direct observation within the system because the system is actually not there, and you interfere with steering the system. When you are driving a car, for example, and you have in your car three psychologists sitting and looking at you, you will definitely behave different to the situation where the psychologists are not within there. So the recommendation is one of the method of choice, direct observation, but you have to add additional methods to analysis human factors, and workload. What you also can do is surveys and questionnaires. Surveys are quick, they are cheap, they are comprehensive. The problem is they are not very accurate and often invalid for the current situation and for determining the workload in the situation, because asking people on a certain amount, what is your workload in the situation, people are not able to understand this question because they have not really access to their own workload. Especially in very stressful situations, people are not able to recognize their own stress level. So surveys and questionnaires are okay for some tasks but not for all. But in combination with first method, direct observation, it produces a better picture. One method we have from software engineering is a cognitive walk-through, is a psychological guided mental simulation of the situation, where you go through the certain steps you have to do, open the car, starting the engine, making the gear, running on the street, and all these things, so step-by-step, you produce a mental walk-through through a situation and try to determine what you have to do, what are the problems you have to do, what is easy to do, what is routinized, what is a problem-solving task and so on. The problem is very accurate and it's a very valid method, but it's a simulation of course, and you have to deal with certain experts who have a good experience with this situation. It is good for the situations or for systems which are in producing or in designing and new, where we have not the possibility to access the system itself. For example, if you are building a new ship or a new car, a new plane, you can make a cognitive walk-through with people and supported by technical things like CAD picture of the system, for example. When we combine all this method, direct observation, surveys, questionnaire, and cognitive walk-through, we typically get a good picture of the workload of people in a certain situation. When we talk about human factors and workload, it seems we are talking about psychology, that is not really true. We're talking about, how people manage a task, how people manage certain workflow. They have a certain input in the situation, which are goals, the task itself, which are some dependencies and some additional external factors, and they have to produce an output, which is information, which is a new task, which is a report with communication or what else. So the interesting point with human factors analysis and determining the workload is not only looking at the person itself, but looking at the person doing a certain task or certain workflow. To understand the influencing factors within human factors environment and better understand workload, you have to take into account certain aspects. One of these aspect is what you can call human factor system design. That is, you have to encounter psychological capabilities and limitations of people interacting with the system. You can use things like thinking tools, support people, or you can have additional sensory abilities to make them act in a certain situation. But the crucial point is you have to understand that people have certain limitations, cognitive limitation, emotional limitation, stress limitation. But they have also capabilities which can use, problem-solving capabilities, intuition, creativity, and all these things. A perfect system design is a system design where we are taking into account capabilities and limitation of the technical system and the capabilities and limitations of the people involved. The next part is human health, psychological and mental stress. A fatigue produces problems, produces unhealthy situations. You have to think about how long will people stay in a stressful situation? What can I do to support some in a fatigue situation, and so else. You have to see an overall system safety approach. System safety does not only mean you have to produce a safe technical system, a safe engine, a safe machine, a safe car, a safe plan, but you have to ensure that the whole system, including possible human errors are safe. We have on one slide, the amount of 70 percent of the causes of an error, or the cause of an accident to human error. So it's a crucial point if it is in our overall system design to reduce human error. Not necessary to substitute human error by machine, but to take into account what errors could occur. You have of course the aspect of training. Training for human performance, training for crew resource management, training for preparedness and awareness for the current situation. But you have to be careful with the topic of training. We quite often see that we have a bad technical system design, and people say, "Okay. Then we have to train the people." You cannot use training as a [inaudible] for a bad system design, a bad system design is bad and you cannot solve it or heal it with a training, you have to reduce the training, only on this aspects on to interact with a good designed system. Acquisition, which is about testing, is about evaluation of system, which is about finding the right ergonomic system, finding the right man-machine interface for the current situation, and of course, all manpower and personnel. Which kind of preparedness do I need? Which kind of expertise do I need? Which kind of amount of people do I need? When I take all these factors into account, all these influence factors on task and workflow, I am on the good way to not reduce workload to zero, of course not, but to reduce the workload and the human factors erroneous to a level which is handleable. So far for today. Thank you very much.
