Human Factor

Safety Analysis:
Design to
Minimize Human
Error

1
 Human Factors

• Key objectives of HF is to
design systems that
• people can use
• increase efficiency and
performance
• minimize the risks of
errors

[Link]
Human factors look at
• Training—Equips the user with the knowledge, skills, and abilities required
for the task
• Personnel (recruiting, retention)—Addresses all aspects of personnel
requirements
• Habitability—Ensures all aspects of the living and working spaces (including
environmental and operational) are designed for the operator in mind
• Engineering design features—Determines the design features of the
equipment that minimize human error and reduce risk of accidents or injury
• Organizational—Identifies how the organization supports the success of the
human operator through its infrastructure
 Conducting Human Factors Safety Analysis

Step 1 Step 2 Step 3 Step 4 Step 5 Step 6

Describe the List and analyze the Analyze the human Screen the Quantify the errors Recommend
system goals and related human errors. identified errors and estimate how changes to the
functions. operations. and decide which they will affect the system that will
Define the system A qualitative tool— ones are worth rest of the system. reduce the impact
Look at how the quantifying. of human errors.
hazards of interest: task analysis—will task can fail, what
system functions be employed. errors can occur, Assess which errors
that may be and how the would have the
influenced by system can recover largest impact on
human errors. It analyzes how the
task is performed from them. the safety of the
Use one of the and what types of system.
other safety tools aids are needed to
(e.g., HAZOP) support
performance.
HUMAN FACTORS
SAFETY ANALYSIS
• Billings and Reynard
(1981) say that 70%–
90% of all system
failures are due to
human error
HUMAN FACTORS
SAFETY ANALYSIS
• Billings and Reynard
(1981) say that 70%–
90% of all system
failures are due to
human error
Human Error

• an error is a failure of achieving the intended outcome in a planned

sequence of mental or physical activities (Reason, 1992)
• an out-of-tolerance action within the human–machine system
• occurs when the human operator is mismatched with the task at
hand
• significant contributor to many accidents
• need to make the safe operation of your systems less dependent on
how well people can operate them
Human Error?
Human Error?
Human Error - Classification

• Various classification schemes exist

ü Discrete action classifications
ü Information processing classifications
[Link] Action Classifications
• One of the simplest (Swain & Guttman 1983)
• Errors of omission - forget to do something
• Errors of commission - doing the task incorrectly
• Sequence errors - out of order
• Timing errors - too slow - too fast - too late
[Link] Processing Classifications
• Rouse & Rouse (1983) propose one scheme
• This scheme follows the information processing assumed to occur
when humans operate and control systems, such as:-
• an aircraft
• a ship
• a power plant
[Link] Processing Classifications
[Link] Processing Classifications
Mistakes
rule Slips
Attention Resources
knowledge

Response Response
Perception
Registration

Selection Execution
Sensory

Decision
Making

Working
Memory
Lapses and
Mode Error
Long-Term Memory
[Link] Processing Classifications
• Attentional Failures
Slips: Right intention incorrectly executed
• Intrusion – masuk ke area/lokasi berbahaya
• Commission – kesalahan dalam melakukan aktivitas
• Reversal – mencoba menghentikan aktivitas yang sudah
berjalan
• Misordering – melakukan aktivitas dalam urutan yang salah
• Mistiming – gagal menjalankan aktivitas dalam waktu yang
ditentukan/sesuai
[Link] Processing Classifications
• Memory Failures
üLapses: failure to carry out an action
• Error of Omission (working memory): Melupakan aktivitas penting

üMode Errors: Making the right response, but while in the wrong
mode of operation
Examples:
leave keyboard in shift mode while trying to type a numeral
driving in wrong gear
[Link] Processing Classifications

• Rule-based Failures
• Misaplikasi aturan yang baik
• Aplikasi aturan buruk

• Knowledge-based Mistakes – wrong solution

because individual did not accurately assess the
situation
Systematic Human Error
Reduction and Prediction
Analysis (SHERPA)
SHERPA
• human-error prediction technique that
analyzes tasks and identifies potential
solutions to errors in a structured
manner
• originally designed to assist people in
the process industries (e.g.,
conventional and nuclear power
generation, petrochemical processing,
oil and gas extraction, and power
distribution)
• 8 - Systematic steps
 Step 1 Hierarchical Task Analysis (HTA)
Analysis of work activities:
task performance can be expressed
in terms of a hierarchy of goals
(what the person is seeking to
achieve), operations (the activities
executed to achieve the goals), and
plans (the sequence in which the
operations are executed)
Step 2: Task Classification
• Action (e.g., pressing a button, pulling a switch, opening a
door)
• Retrieval (e.g., getting information from a screen or manual
• Checking (e.g., conducting a procedural check)
• Selection (e.g., choosing one alternative over another)
• Information communication (e.g., talking to another party)
 Step 3: Human-Error Identification (HEI)

Consider
credible
error modes
associated
with each
activity, using
the error
taxonomy
• Step 4: Consequence Analysis: Considering the consequence of each error on a
system is an essential next step, as the consequence has implications for the criticality of the error.

• Step 5: Recovery Analysis: If there is a later task step at which the error could be
recovered, it is entered next. If there is no recovery step, then "None" is entered.

• Step 6: Ordinal Probability Analysis: An ordinal probability value is entered as

either low, medium, or high. If the error has never been known to occur, then a low (L) probability is
assigned. If the error has occurred on previous occasions, then a medium (M) probability is assigned.
Finally, if the error occurs frequently, then a high (H) probability is assigned. The assigned
classification relies upon historical data and/or a subject-matter expert.

• Step 7: Criticality Analysis: Criticality is assigned in a binary manner. If the error

would lead to a serious incident (this would have to be defined clearly before the analysis), then it is
labeled as critical (denoted thus: !). Typically, a critical consequence would be one that would lead to
substantial damage to plant or product and/or injury to personnel.
Step 8: Remedy Analysis
• changes to the work system that could have prevented the error from occurring or, at the very least, reduced
the consequences.
• a structured brainstorming exercise to propose ways of circumventing the error or to reduce the effects of the
error
1. Equipment (e.g., redesign or modification of existing equipment)
2. Training (e.g., changes in training provided)
3. Procedures (e.g., provision of new, or redesign of old, procedures)
4. Organizational (e.g., changes in organizational policy or culture)

Each recommendation is analyzed with respect to four criteria:

1. Incident prevention efficacy (the degree to which the recommendation, if implemented, would prevent
the incident from occurring)
2. Cost effectiveness (the ratio of the cost of implementing the recommendation to the cost of the incident
× the expected incident frequency)
3. User acceptance(the degree to which workers and organization are likely to accept the implementation of
the recommendation)
4. Practicability (the technical and social feasibility of recommendation)
Advantages and Disanvantages
ADV:
• Structured and comprehensive procedure, yet maintains usability
• Taxonomy prompts analyst for potential errors
• Encouraging validity and reliability data
• Substantial time economy compared with observation
• Error reduction strategies offered as part of the analysis, in addition to predicted errors
DIS:
• Can be tedious and time-consuming for complex tasks.
• Extra work is involved if HTA is not already available.
• Does not model cognitive components of error mechanisms.
• Some predicted errors and remedies are unlikely or lack credibility, thus posing a false economy.
• Current taxonomy lacks generalizability.