Human Machine System 1 Flashcards
(28 cards)
Blunt end - Sharp end
Manufacturer - Operator
Automation
Technology by which a process or procedure is performed without human assistance. For example using control systems to operate machinery in a factory.
Pros of automation
Fast, precise, quality
Cons of automation
Out-of-the-loop performance problems, loss of skills, inappropriate trust
Cons of automation
Out-of-the-loop performance problems, loss of skills, inappropriate trust (over-trust, distrust, expectations/capabilities)
Ironies of automation
The more automation, the more dependent you are of the people left in the “system”
Work will only take another form
Vigilance during supervision
Workload higher sometimes (malfunctions)
Risk of human error is not removed
Deterioration of knowledge
Additional work is created
More difficult work (operator doesn’t know how the task is being done)
Out-of-the-loop
Performance problem that leaves the operator unable to take over manual control in case of failure (vigilance, passive info processing)
Principles of function allocation: “The left over principle”
Human takes care of the left over tasks.
- Task occur seldom
- Cost of automation > benefit
- Too complex to do
Operator is left with difficult and unusual tasks
Principles of function allocation: “The compensatory principle”
Human and machine compensate each other.
Use of “fitts list” / MABA-MABA
(Machines are better at - men are better at)
Human strengths: flexible, creative, problem solving, advanced decisions, differentiate signals from noise
Machine strengths: fast, accurate, power, repetition, hostile environments, not bored/tired, no breaks, fast info processing
Principles of function allocation: “The complementary princible”
Human and machine complement each other.
- Adapt automation to operator needs
- Automation as a partner
- Robust system when function allocation overlapps
- Collaboration rather than interaction
What to consider when level of automation increases
- How to keep operator in the loop?
- See allocation of functions as a complementing system, not competition.
- Operator work is realistic and considers human limitations.
- Meaningful operator work and not only the left over parts.
- Operator can’t take over system they are not involved in.
How to design better automation?
- System transparency, support understanding
- Use automation as support to optimise system performance
- Use knowledge of human cognition in design
- Consider the operator competence and knowledge
- Apply systems view
Examples of bad desingn
- learn by bad example
- inconsistency
- mixture of text and symbols
- illogical sequence
- conflicting messages
- unclear cues / bad cues
- offend against conventions
- complicated use
- difference btwn alternatives
- too much information
- overexplicit
Basic principles when designing information elements
- Right info, in a right way and at a right time
How different types of information should be presented?
Qualitative:
- information about status
- indicators should be easy to distinguish
- no sound = normal, sound = warning
Quantitative:
- numerical
- figures: plain and clear to read (font, size)
- scales: plain and clear, one per tool/level
- pointer should not cover figures
Design aspects when designing a scale
- Use clockwise movement
- Use only linear scales
- Pointer shouldn’t cover the figures/lines but be close to them
- Use 1/2 or 1/5 subvision
- Avoid using multiplication
- Required accuracy > division btwn values/scale
- tip should be narrower than the lines
Digital (numeric) vs analogue (graphical)
High precision - Changes, fast reading, deviations
Text vs symbols
Text:
- no need for extra interpretation
Symbol:
- intuitive/familiar ones without text
- shorter reaction time
(- context?)
Use audible information when… (same aspects apply to haptic information)
...supplementing visual information ...when vision is already busy ...attract attention ...as a indicator ...when transferring information
Visual vs audible
Visual:
- info high in complexity
- info low in priority
Audible:
- info low in complexity
- info high in priority
General designing principles related to information
Detection:
- Good visibility, consider surrounding equipment
Recognition:
- Colours, text fonts,…, split info to meaningful groups
Understanding:
- No long words, no uncommon expressions, no complicated symbols and icons
- The user needs to have sufficient knowledge to interpret the presented info
Overall design principles for designing controls
- Clear mapping btwn layout and functions
- Avoid spatial transformations
- Order of elements should be same in different panels
- Group elements with common functions
- Use redundancy for important information
- Position text and symbol labels ABOVE the elements
- Control elements should be on the right side of the information
- Most often used elements closest to the user
- Related info and control elements positioned together
Coding princibles
- Coding by location: control near function - powerful principle
- Coding by colour: different colour for functions (e. g. red-stop)
- Coding by size: Three different sizes > operator see/feel difference
- Coding by shape: Control has shape of the function - landing gear
- Coding by labelling: Written description for a function
- Coding by mode of operation: Controls have different feel to them - e.g. brake and accelerator
Mental models
Mental representations of how object works and how interface controls affect it.
People have preconceived models that you might not be able to change. User interface should match the mental model or clearly tell that the old models don’t work.
