During the design phase of any process, safety risks are assessed formally using procedures such
1 Overview
During the design phase of any process, safety risks are assessed formally using procedures such
as HAZOP. Quantitative methods such as Fault Tree, LOPA often follow once the components
identified through the HAZOP are in place. Organisations will usually approach HAZOP in a similar
way, with the exact details of the HAZOP, the way in which it is recorded may differ. Fault trees
are usually adopted across several sectors in order to assess probability and reliability of a system
comprising of a number of components.
2 Assignment format and Deadline
This assignment measures your comprehension of both HAZOP and fault trees.
You will produce a process risk assessment report by performing a) a HAZOP analysis on a process
and b) a fault tree analysis of a process or piece of equipment from the (industrial) sector in
which you work.
3 Requirements of the written report
The report shall have two sections addressing the following topics:
Hazop Analysis
Fault Tree Analysis
3.1 HAZOP Analysis
Perform a HAZOP analysis on the following information shall be produced:
Process Summary in your own words.
Scope and objective of the HAZOP analysis
State the original design intentions with details of process data (i.e., flow rates,
temperature, pressure etc,)
Generate a suitable HAZOP sheet to record your deliberations. Populate with the range
of guidewords details found in the course notes. You may use a one that is used in your
workplace.
Identify the major recommendations you would make to the P&ID as given.
HAZOP studies would normally be conducted by a team with appropriate skills. However, this
exercise is to ensure you are competent with the procedure.
3.2 Fault Tree Analysis
Select a process or piece of equipment from the industrial sector in which you work to perform
a fault tree analysis following these steps:
i. Define the top event
ii. Build the Fault tree
iii. Identify the Cut sets
As minimum, the following information shall be produced:
Process or equipment description
Define the top event and provide a detailed description for it.
Build the fault tree and provide a detailed explanation of the fault tree diagram. The fault
tree shall be built down to at least a second-level contributor of events (Refer to appendix
3.3 for an example).
Assign either failure probabilities or failure rate on demand to the primary events and
events not developed further and indicate their data source. However, failure rate on
demand shall be assigned to at least 40% of the above events. Failure rate on demand
and probabilities shall be either selected from company maintenance records (historical
data, if available), vendor records, third-party databases (if accessible), or using data from
the document indicated in the appendix. If, even after an extensive search no data is
available then data can be assumed.
It can be assumed that events, per level contributor of events, are independent from each other
(example, refer to appendix 3)
Identify the minimum cut sets and their probabilities.
Determine the likelihood of occurrence of the Top Event.
o What is the greatest minimum cut set contributor to this probability?
o And what is your field experience related to this finding?
4 Appendix
4.1 Oil Storage Tank HAZOP
A mild steel cylindrical tank of capacity 1,000 m3 is used for the storage of heavy oil. The tank,
illustrated below, has a fixed roof with three vent pipes, each of which is fitted with a flame trap.
Oil is transferred in and out of the tank at 120oC because it would be too viscous to handle at
ambient temperature. The tank has an internal steam coil which can be used to keep the
contents hot, saturated steam being available at 5 bar and 151.8oC. The pump used to transfer
the oil is a standard centrifugal type, which can be used to either fill or empty the tank by setting
valves to direct the oil flow as described below.
Notes:
1- The tank was designed and has been utilised for sporadic operations of at least once every
three months. However, lately the tank started to be utilised for more frequent oil storage
operations of at least once every week. It is envisioned that this operation mode will
continue due to operational constraints. This has led to certain operational incidents, reason
for which the HAZOP study has been proposed.
2- To fill the tank, open valves 1 and 3 and close valves 2 and 4; valve 3 can be used to regulate
flow. To empty the tank, open valves 2 and 4 and close valves 1 and 3; valve 2 can be used
to regulate flow.
3- The tank has only instrumentation for local indication such as level and temperature and it
does not have any additional physical protection other than the flame arrestors/traps.
4- The tank is within a tank farm with a Central Control Room manned 24-hour a day.
5- The tank farm is controlled and monitored by a DCS.
6- The plant has a ESD designed to shut down and/or isolate process equipment or shutdown
the plant in case of emergency conditions.
Perform a HAZOP analysis for filling the tank with hot oil.
4.2 Failure Rate and Event Data - Health and Safety Execute
(HSE), Land Use Permit (LUP)
The failure rates quoted within the document indicated in the link below were derived and are
intended for use on Land Use Planning cases around major hazard sites and pipelines and on
applications for hazardous substances consent. They were NOT originally intended for use in
COMAH Safety Report Assessment because they do not necessarily take account of all factors
that could be relevant and significant at particular installations. However, in the absence of sitespecificdata, the values given there may serve as a starting point for safety reports.
https://www.hse.gov.uk/landuseplanning/failure-rates.pdfAlternatively see Lees, Loss Prevention Series
4.3 Fault Tree - Second-level structure
4.4 Independent events at contributor level
Failure probability
Failure Rate
Failure function (F=1-R) shall be used to determine failure probabilities when failure rates
on demand are given. Function F is the cumulative probability of failure at time t.
