Hunting for ways to reduce design-induced end user error

by Yvonne Toft

Introduction · Questions · Web Resources · The Big Question · Conclusion · Rubric · Guide

Introduction

Human error is the cause of most accidents. Listen to any news broadcast following any major incident and you will have this ‘fact’ reinforced over and over. Researchers have actually found that 80 – 90% of all industrial accidents are caused by human error. But did you know that there is at least two types of error?

1. Active errors (those that occur at the front end of an operation ie. end users – the effects are felt almost immediately, for example, in the Chernobyl nuclear accident the front line operators removed too many control rods and cut water from the turbines)

2. Latent errors (those errors mostly made by those that are removed from the direct control interface by space and time, they are the designers, high level decision makers, managers, maintenance and construction personnel; the errors may lay dormant in a system for many years, the consequence only realised when they combine with other factors to breach a system’s defences, for example, in the Waterfall train crash the ‘dead man’ system had a design flaw that was always going to cause an accident – the only unknown was when!)

Next time you hear about an accident being caused by ‘human error’ - stop for one moment and ask yourself – Which human??

Many people claim that we live in an Information Age, where trying to learn amidst the explosion of facts and opinions is like sipping water from a fire hose. And the Internet seems to be the world's biggest fire hydrant. Still, just as we need water to survive, scholars need information to learn. Your task in this activity is to answer some key questions on the topic. The Web offers interesting facts and opinions that may not be available anywhere else, but to find the truth you may have to read between the lines. If you want an idea of how your learning will be evaluated, read this evaluation rubric. Finally, remember, this activity is about thinking and learning, not copying and pasting.

Questions

  1. What area of ergonomics helps us understand the way people think and respond to the world around them?
  2. There are three main stages at which information is transformed. Imagine that you are observing a driver who has pulled up at traffic lights. What is happening at each of these three stages as the driver decides what to do next?
  3. You have been asked to assist investigators of an industrial accident (a traffic accident that has been complicated by a chemical spillage). How can each of your senses keep you out of danger?
  4. Imagine that you notice the sound of the radio in the other room, how do you notice when it becomes louder? How could this knowledge help you understand why an end user of a system might make a mistake in responding to an emergency alarm?
  5. The NORAD line was designed to provide early warning of a nuclear missile attack by the Soviet Union. Powerful radar beams swept across the stark arctic terrain, constantly probing for enemy missiles. One day, radar screens lit up with hundreds of blips. Was this a surprise attack on the part of the USSR? If so, the US had less than 30 minutes before the first bombs began to explode...? What two variables would influence the observer's reactions?
  6. What theory can help us understand why we attempt to organize information into certain groups and interpret the information completely without unneeded repetition. For example, when you see one dot, you perceive it as such, but when you see five dots together, you group them together by saying a 'row of dots.' Without this tendency to group our perceptions, that same row would be seen as 'dot, dot, dot, dot, dot,' taking both longer to process and reducing our perceptive ability. How could this tendency to group our perceptions lead to end user error?
  7. What type of memory do we target when training a crane operator? If we think of the crane driver as an information processing system - explain what is occuring during the training session?
  8. 'In a dual task study, performance while doing one task is compared to performance doing two tasks. If performance with two tasks is worse than one alone, then it can be concluded that the tasks use the same resources. However, if performance with two tasks does not differ from performance of one alone, then it can be concluded that the tasks do not use the same resources, or that one of the two tasks is automatic (automatic processes are said to not use any resources).' What resources is the author referring to and what are the implications for error by end users of a system?
  9. Memory is critical to incident recall, features of an event will determine the way that an event is recalled. What factors might influence the way you recall information about an incident that you have witnessed?
  10. The NORAD line was designed to provide early warning of a nuclear missile attack by the Soviet Union. Powerful radar beams swept across the stark arctic terrain, constantly probing for enemy missiles. One day, radar screens lit up with hundreds of blips. Was this a surprise attack on the part of the USSR? If so, the US had less than 30 minutes before the first bombs began to explode...? How could we improve performance of these operators appropriate to the type of attention task?
  11. The National Transport Safety Bureau published a set of 37 'crew-caused' accidents that involved 'tactical decision errors'. A common pattern was the crew's decision to continue with their original plan when conditions suggested that other courses of action might be more prudent. In other words, they decided to 'go' in a 'no go' situation, usually in the face of ambiguous or dynamically changing conditions, e.g. continuing with a landing when it might have been more appropriate to go-around. Four factors were hypothesized as possible contributors to these decision errors. What are these four factors and how could these same four factors cause design-induced end user error?
  12. What design principle can be employed to ensure that an operator looking at a bank of displays can easily see if one parameter is outside of normal operation?
  13. What is the best type of control for an end user who is required to make (a) a precise adjustment; e.g., radio volume; and (b) tactile feedback? Imagine a situation when tactile feedback was important but not included in the design. Describe the situation and the consequences.
  14. What configuration of control burner (stove top) arrangements lead to 10% of users making an error? What design principles could we apply to the design of stove tops to reduce end user error?
  15. What error might a railroad worker make as a result of the Müller-Lyer effect?
  16. What are the functions of a graphic symbol? How could you reduce end user error in a sign directing traffic to an airport?
  17. What type of dynamic display should be used when rate of change is important (for example a power house control panel) and why?
  18. In what type of situations would we use an auditory display rather than a visual display? How could we maximise the effectiveness of an auditory display in a control room?
  19. What sort of display could we use to improve car sales and why?
  20. What are the potential consequences of a worker's experience of 'subjective workload'?
  21. Murphy, a US Air Force officer at Edwards Air Force Base in California, was quoted in the 1950's as saying: 'If something can go wrong, it will.' And Isaac Newton, of course, said: 'For every action, there is an equal and opposite reaction.' What can we learn from Murphy, Newton and Rocket Technician Anders Svanson to prevent future accidents?

Web Resources Links

Auditory Displays
Perception
Static Displays
Sensation
Controls
Signal Detection Theory
Long Term Memory
The Senses
Human Memory
Mental Workload
International Ergonomics Association
Dynamic Displays
Short Term Memory
Decision Making
Gagné's Information Processing Model
Displaying and Coding Information
Compatibility Relationships
Disaster by Design
Other Types of Displays
Attention Resources
Process of Seeing


The Big Question

Your challenge is to come up with an answer to the question: How can ergonomics knowledge be used in design to reduce end user error (that is - reduce the possibility of latent error contributing to active errors!)? Consider what you learned about the way people perceive, process and respond to information from the systems they interact with.



Conclusion

Your journey presented many possibilities for improving designs. People used to think that learning was all about gaining information and memorizing facts. We hope you've learned some important information in this Knowledge Hunt, but more than that, it would be great if you saw ways to use this knowledge to expand your understanding of the topic. If you want to check how you've done, read the evaluation rubric for this activity. Remember, with so much info exploding all around us, finding a nugget of knowledge is not as important as doing something with it. What will you do with your learning?


Web and Flow, by ozline.com created by Yvonne Toft
email: y.toft@cqu.edu.au
http://web-and-flow.com/members/ytoft/folder4/hunt.htm