Many system reliability predictive methods are based solely on equipment failures, neglecting the human component of man–machine systems (MMS). These methods do not consider the identification of the root causes of human errors. Accelerating technological development leads to an increased importance of safety aspects for organizations as well as for their environment. Therefore, especially in the case of high hazard organizations an expanded view of safety – system safety including human factors is needed. These organizations need appropriate structures as well as rules for the treatment of safety relevant actions or tasks. The system safety approach is reflected in the recent developmental stage in safety research, which started with a focus on technology and its extension to human errors, socio-technical systems and recently to the inter-organizational perspective. Accident causation theories as well as approaches to organizational learning are the theoretical background. Nevertheless, the majority of measurements (methods) and interventions remain in the former stages, i.e. technical or human error orientation. This problem will be discussed by the means of examples. The contribution will end with an outlook to possible future ways of integrating the new developments in safety research.

1. SEMINAR
ON
“HUMAN ELEMENT FACTOR AFFECTING
RELIABLITY & SAFETY-Treating Human errors
and risk factors in probabilistic analysis”
BY
DUSHYANT KALCHURI
M.TECH(PRODUCTION ENGG.)
DEPARTMENT OF
MECHANICAL ENGINEERING

2. CONTENTS :
• Introduction
• Faulty measures undertaken during the handling of equipment
And Unsafe Acts
• Case Study On:
Computer System Interruptions By Human Errors
Utility Distributions Interruptions By Human Errors
• Classification of Human Error according to system orientation
• Human Reliability analysis
• Technique for Human Error Rate Prediction
• Error Prevention / Remediation
• Accident Injury Sequence Model
• Safety analysis
• References

3. INTRODUCTION:
• Many system reliability predictive methods are based
solely on equipment failures, neglecting the human
component of man–machine systems (MMS).
• The reliability and safety of industrial and commercial
power systems and processes (i.e., MMS) are dependent
upon human characteristics and many dependent and
dynamic interactive factors .
• The consequences of human errors are very diverse and
can range from damage to equipment and property, injury to
personnel or fatalities, to disruption of scheduled system
operation, all of which represent a significant cost to society.

4. HUMAN ERROR:
• A failure on the part of the human to perform a
prescribed act or task within specified limits of
accuracy, sequence, or time, which could result in
damage to equipment and property and disruption of
scheduled operations or have no consequences at all.
• “ most of the human errors occur because humans are
capable of doing so many different things in many
diverse ways.”
• Generally 20%–50% of all equipment failures are
due to human errors.

5. WHY A HUMAN PERFORMANCE
IMPROVEMENT APPROACH??
80% Human Error 30%
Individual
20% Equipment
Failures
Human Error
Unwanted Outcomes
70% Latent
Organization
Weaknesses

6. INDUSTRY EVENT CAUSES
DUE TO HUMAN PERFORMANCE/ERROR
Source: INPO, Event Database, March 2000. For all events during 1998 and 1999.
215
26 39
88
192
654
9 20
160
82
806
73
118
0
100
200
300
400
500
600
700
800
900
C
hange
M
anagem
ent
Environm
entalC
onditions
Hum
an-m
achine
Interface
Supervisory
M
ethods
W
ork
O
rganization/Planning
W
ritten
Procedure
R
esource
M
anagem
ent
W
ork
Schedule
Training/Q
ualification
VerbalC
om
m
unicationsW
ork
Practices
M
anagerialM
ethodsO
ther/Unknown
NumberofCauses
1,676 = Org behavior (68%)
806 = Individual behavior (32%)

7. TAXONOMY OF HUMAN ERROR:
Interpretation
Situation
Assessment
Plan
Intention of
Action
Action
Execution
Stimulus
Evidence
Memory
MISTAKES SLIPS
LAPSES &
MODE ERRORS
Knowledge Rule

8. TAXONOMY OF HUMAN ERROR
MISAKES:
• Mistakes – failure to come up with appropriate solution
• Takes place at level of perception, memory, or
cognition
• Knowledge-based Mistakes – wrong solution because
individual did not accurately assess the situation.
• Caused by poor heuristics/biases, insufficient info,
info overload
• Rule-based Mistakes – invoking wrong rule for given
situation
• Often made with confidence

9. TAXONOMY OF HUMAN ERROR
SLIPS:
• Slips – Right intention incorrectly executed (oops!)
• Capture errors – similar situation elicits action, which
may be wrong in “this” situation. Likely to result when:
• Intended action is similar to routine behavior
• Hitting enter key when software asks, “sure you
want to exit without saving?”
• Either stimulus or response is related to incorrect
response
• Hit “3” instead of “#” on phone to hear next
message, because “3” is what I hit to hear the first
message

10. TAXONOMY OF HUMAN ERROR
LAPSES & MODE ERRORS:
• Lapses – failure to carry out an action
• Error of Omission (working memory)
– Examples: Forgetting to close gas cap, failure to
put safety on before cleaning gun, failure to
remove objects from surgical patient
• Mode 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,
going wrong direction because display was north-
up when thought it was nose-up

11. FAULTY MEASURES UNDERTAKEN
DURING THE HANDLING OF
EQUIPMENTS:
1. Loose connections;
2. Faulty installation;
3. Improper grounding;
4. Defective parts;
5. Ground faults in equipment;
6. Unguarded live parts.
These Conditions lead to plant interruptions and
disruption of processes and degrades Reliability.

12. • According to many safety and health laws, employers
must provide a workplace where workers will not be
exposed to hazards, where practicable.
• Workers must receive training, instruction,
supervision, and information so they are not exposed
to hazards.

13. Examples of FAULTY/ UNSAFE acts:
1. Failure to de-energize, Lockout and Tag-Out hazards
during maintenance, repair, or inspections;
2. Use of defective and unsafe tools;
3. Use of tools or equipment too close to energized
parts;
4. Etc..

14. CASE STUDIES ON THE
FREQUENCY OF HUMAN ERRORS:
• Computer System Interruptions Caused
By Human Errors
• Utility Distributions Interruptions By
Human Errors

15. COMPUTER SYSTEM INTERRUPTION
CAUSED BY HUMAN ERROR:
• A ten-year study at the University of Alberta’s central
computer system was conducted analyzing the
frequency of computer system interruptions caused
by operator errors.
• A human or computer operator error is defined as an
act or set of acts which results in a computer system
interruption, and the system is restored to an
operational state either by initial program loading or
restarting.

16. • The computer system runs continuously 24 hours a
day, except for maintenance periods early in the
mornings on the weekends to minimize the impact of
the scheduled interruptions on the users.
• The annual number of computer system interruptions
caused by operator errors is shown in Fig (Next
Slide).

17. YEAR OF STUDY:

18. • The total number of computer system interruptions
caused by operator errors per year averaged
approximately 25.
• Table 1 reveals the various percentages of
interruptions attributed to the primary causes of
computer system interruptions in which operators
errors accounted 7.4% of computer system
interruptions.

20. 1. DAY OF THE WEEK OF COMPUTER
SYSTEM INTERRUPTION:
• A ten year study of the average frequency of computer
system interruptions per given day of the week
confirmed belief of operators, as is shown in Fig. 2.
• The average frequency of computer interruptions was
higher during the “weekdays” (i.e., Monday through
Friday) than on the weekends, when the system loading
was reduced.
• This supported the operators belief that “weekdays” were
more prone to computer system interruptions than
Saturday and Sunday.

21. DAY OF WEEK

22. 2. TIME OF THE DAYOF COMPUTER
SYSTEM INTERRUPTION:
• Many users of the computer system claimed that there
appeared to be more interruptions in the morning than
during the remainder of the day.
• The loading on the system peaked between 8–9 a.m.,
and the load dropped off between 4–5 p.m. and remained
fairly steady for the remaining time periods.
• It is clear that the sudden increase in computer system
loading and operator stress between 8–9 a.m. was
directly correlated with a significant increase in the
frequency of operator errors resulting in computer
system interruptions.

23. TIME OF DAY

24. UTILITY DISTRIBUTION INTERRUPTION
CAUSED BY HUMAN ERROR:
• The electric utility distribution system customer
interruptions were recorded for the past 30 years by
the Canadian Electricity Association (CEA) in
Canada.
• It can be seen that the human element accounts for
approximately 1.7% of the total number of
distribution system interruptions.
• Other factors, such as scheduled outages, lightning,
and defective equipment were the dominant causes of
distribution system interruptions.

25. ELECTRIC UTILITY LOST TIME
DUE TO INJURY ACCIDENTS:
• To measure the impact of injury accidents on
productivity in terms of hours in the workplace, the
CEA uses an index called the severity rate.
• The severity rate equals the number of calendar days
lost due to injury accidents per millions of hours
worked. Typical rates are shown in Fig. 5
• The severity rate remains fairly constant for several
years, averaging about 500 days lost per million
hours worked.

26. SEVERITY RATE

27. CLASSIFICATION OF HUMAN
ERRORS ACCORDING TO SYSTEM
ORIENTATION:
• Human errors can occur at any stage in the life
of a system.
• It occurs from the original design inadequacies,
to installation deficiencies and operating and
maintenance human anomalies.

28. Continued…
Design
Installation
Assembly
Inspection
Operating
Maintenance
Classification
Of Human
Error

29. DESIGN ERROR:
• It can be attributed to the physical structure of a system
with basically the following three types of
inadequacies:
1. failure to implement human needs in the design.
2. assigning inappropriate functions to persons, e.g., lack
of definition of primary work tasks;
3. failure to ensure the effectiveness of the man and
machine component interactions.

30. • INSTALLATION ERROR:
• This are primarily due to the failure to install
equipment by humans according to instructions or
blueprints, assuming these drawings are correct, and
poor workmanship when operating under severe time
constraints.
• The inspection criteria of evaluation is dependent
upon the inspector’s knowledge of the system and the
relation between its interacting parts.
• According to study an average inspection
effectiveness is close to 85%.
• INSPECTION ERRORS:

31. ASSEMBLY ERROR:
• This errors are the result of poor workmanship.
These errors are often discovered after the
installation process when they disrupt scheduled
system operations.
• Examples are:
1) use of incorrect component;
2) use of incorrect tools;
3) omitting a component;
4) improper connections;
5) improper handling of equipment.

32. OPERATION ERRORS:
• This error is subject to human operating errors.
Situations that lead to these errors are as follows:
1) lack of proper procedures;
2) task complexity and overload conditions;
3) poor personnel selection and training;
4) operator carelessness and lack of interest;
5) poor environmental conditions.

33. MAINTENANCE ERROR:
• This errors are primarily due to the incorrect
repair/replacement/service activities of equipment.
• Examples of maintenance errors are the following:
1) incorrect calibration of instruments, e.g., relays,
computer controls, and sensors;
2) failure to follow maintenance schedules and
procedures;
3) incorrect equipment cleaning procedures.

34. HUMAN RELIABILITY ANALYSIS
• Human Reliability Analysis – predict reliability
of system in terms of probability of failure or mean
time between failures (MTBF) when system is
designed to work in parallel or series
.9 .9
.9
.9
Series
Parallel
Reliability = .9 x .9 = .81
P(failure) = 1 - .81 = .19
Reliability = 1 – [(1 - .9) (1 - .9)]
= 1 - .01 = .99
P(failure) = 1 - .99 = .01

35. TECHNIQUE FOR HUMAN ERROR
RATE PREDICTION (THERP)
THERP components
1. Human Error Probability
• Ratio of errors made to possible errors
2. Event Tree
• Diagram showing sequence of events
• Probability of success or failure for each
component
3. Other Moderating Factors
• May add in multiplier to account for variables such
as experience level, time, stress, etc.

36. THERP EVENT TREE
a A
ba Ba
S
S
bA BA
F
S
F
S
F
F
Series
Parallel
Series:
P[S] = a(ba)
P[F] = 1 – a(ba) = a(Ba) + A(bA) +
A(BA)
Parallel:
P[S] = 1 – A(BA) = a(ba) + a(Ba) + A(bA)
P[F] = A(BA)
P(successful task B given A)
P(unsuccessful task B given A)
P(success of task B given a)
P(Unsuccessful task B given a)
P(successful task A) P(unsuccessful task A)
Task A = first task
Task B = second task

37. ERROR PREVENTION / REMEDIATION
1. Task Design – design tasks with working memory capacity
in mind
2. Equipment Design
a) Minimize perceptual confusions – ease of
discrimination
• Ex: airplane controls that feel like what they do
(flaps, wheels)
b) Make consequences of action visible – immediate
feedback
• Ex: preview window in some software programs
c) Lockouts – design to prevent wrong actions
• Ex: car that will not let you lock door from outside
without key
d) Reminders – compensate for memory failures
• Ex: ATM reminds you to take your card

38. ERROR PREVENTION / REMEDIATION
(Cont.….)
3. Training – provide opportunity for mistakes in
training, so can learn from them
• Ex: Simulation
4. Assists and Rules – checklists to follow
• Ex: Pilot pre-flight checklist
5. Error-tolerant systems – system allows for error
correction or takes over when operator makes serious
error
• Ex: Undo button

39. ACCIDENT-INJURY SEQUENCE MODEL :
• This Model provide a framework for identifying
the possible root cause of electrical accidents.
• This Model provide a basis for developing
accidents prevention and injury control strategies to
minimize
1. impact of disruption to system operation.
2. occurrences of injuries.

41. SAFETY ANALYSIS
Sequence for identifying potential hazards and recommendations for hazard
reduction: (Weinstein et al. 1978)
1. Task Analysis – How will product be used?
2. Environment Analysis – Where will product be used?
3. User Analysis – Who will use product?
4. Hazard Identification – What is likelihood of hazard
with product?
5. Generate Methods for Hazard Control – What might
eliminate hazards?
6. Evaluate Alternatives – How will alternative designs
affect product performance?
7. Select Hazard Control – Given alternatives, what is best
design to minimize hazards?

42. ACCIDENT INVESTIGATION LEVELS
OF CAUSES
Management Safety Policy & Decisions
Personal Factors
Environmental factors
Unsafe Act Unsafe
Condition
Unplanned Release of Energy
And/or
Hazardous Material
ACCIDENT
Personal Injury
Property Damage
BASIC
CAUSES
INDIRECT
CAUSES
(SYMPTOMS)
DIRECT
CAUSES

43. SAFETY PROGRAMS
1. Identify risks to the company
 identify hazards, hazard controls, accident
frequency, & company losses due to
accidents/incident claims
2. Implement safety programs, includes:
 management involvement, accident
investigation, recommendations for equipment,
safety rules, personal protective equipment,
employee training, safety promotion
3. Measuring program effectiveness
 evaluated by assessing changes in safety
behaviors, accident/incident rates, number of
injuries or death, and number of days off due to
injury

44. CONCLUSION / SUGGESTION:
• Risk-Taking as a Decision Process
• People must know a hazard exists, know what actions
are available, & know the consequences of the safe
behavior vs. alternative behaviors
• Written Warnings and Warning Labels
• Accurately convey the hazards of a product
• Should include a signal word, info pertaining to the
hazard, consequences, & necessary behavior
• Danger: Immediate hazard likely results in severe
injury
• Warning: Hazard could result in injury
• Caution: Hazard or unsafe use my result in minor
injury

45. REFERENCES:
• Human Element Factors Affecting Reliability and Safety, Don O.
Koval and H. Landis Floyd, IEEE Transactions on Industry
Applications, Vol. 34.
• IRACST- International Journal of Research in Management &
Technology (IJRMT), ISSN: 2249-9563 ,Vol. 2, No. 1, 2012,Human
Reliability Analysis: A review of the state of the art
• 4th European-American Workshop on Reliability of NDE -
Th.4.A.1, Integrating Human Factors in Safety and Reliability
Approaches by Babette FAHLBRUCH, TÜV NORD SysTec, Berlin,
Germany http://www.ndt.net/index.php?id=8338

46. Thank You!!!