HAZOP

1
HAZOP

• System Safety HAZOP and Software HAZOP, by Felix
  Redmill, Morris Chudleigh, James Catmur, John
  Wiley Sons, 1999

2
What is HAZOP?

• Technique for identifying and analyzing the
  hazards and operational concerns of a system.
• Central activity a methodical investigation of
  a system description (design representation).

3
What this presentation does not cover

• The book puts a LOT of emphasis on
• Selecting the study initiator
• Selecting the study leader
• Planning the study
• Roles during the study
• Questions vs. follow-up
• Completion criteria
• (P.S. It also tells how to conduct the study
  itself -)

4
Reasonable Limits for this class

• This is a human-intensive activity
• As such, the details on the previous page are of
  extreme importance authors are experienced and
  therefore recognize this
• You wont be able to conduct a HAZOP study on the
  basis of these slides
• Goal Understand what it is set the bar higher

5
Study process itself in a nutshell
Introductions
Presentation of design notation
Examine design methodically one unit at a time
Is it possible to deviate from design intent here?
YES
Examine both consequences and causes of the
possible deviation
NO
NO
Document results
Define follow-up work
YES
Time up?
Agree on documentation
Sign off
6
Examine design methodically each unit in turn

• Suppose the design representation is a collection
  of state transition tables
• Units are states, transitions, event/action pairs
• For EACH, list the recommended attributes (see
  table from the Hazop book)
• For each attribute, use the guide words to
  trigger the questions about ways to deviate

7
The suggested guide words

• No negation of design intention no part of
  design intention is achieved but nothing else
  happens
• More Quantitative increase
• Less Quantitative decrease
• As well as Qualitative increase where all design
  intention is achieved plus additional activity
• Part of Qualitative decrease where only part of
  the design intention is achieved
• Reverse logical opposite of the intention
• Other than complete substituion, where no part
  of the original intention is achieved but
  something quite different happens

8
When timing matters

• Add the following guide words
• Early something happens earlier in time than
  intended
• Late something happens later in time than
  intended
• Before something happens earlier in a sequence
  than intended
• After something happens later in a sequence
  than intended

9
Guide words chosen

• Match the system being examined to appropriate
  table or modify the closest
• Match the design representation
• Note not all guide words apply to all attributes
• For attribute speed of an electric motor, omit
  guide word as well as and part of
• For attribute data flow on a dfd, less is not
  used because meaning covered by part of
• Generally, study leader selects from the guide
  words, provides interpretations based on chosen
  design representation and context, distributes to
  team in advance of the study

10
Applications

• Originally developed for chemical plants
• Book has detailed examples for
• Software using data flow diagrams
• Software using state transition diagrams
• Includes timing attributes of response time and
  repetition time
• Software using various OO models
• Digital electronics
• Communication systems
• Electromechanical systems
• Same guide words, different interpretations

11
See book excerpts

• More detailed outline of the HAZOP process
  Figure 9.2
• For all entities
• For all attributes
• For each guide word
• Is deviation credible?
• Example matrices

12
Fig 9.2
HAZOP meeting process
13
(No Transcript)