15/06/2016

RBI TRAINING COURSE

INTRODUCTION TO RBI
RBI Training Course
Module 01a

DNV GL © 2014 SAFER, SMARTER, GREENER

Scope of the Training

Introduction to RBI
RBI Methodology Theory + with hands on exercises
– Likelihood calculation
– Consequence calculation
Case Study with the RBI software: refinery unit
– Data preparation
– Screening analysis
– Detailed analysis
– Inspection Planning

DNV GL © 2014

1
 15/06/2016

RBI TRAINING COURSE

Agenda

Synergi Plant RBI Onshore - Training Program. Breaks assumed during the day but not shown. Timing approximate.
Module
Item Start End Objectives
Day 1
1 Introductions - Installations, Introduction to Synergi RBI 09:00 12:30
Lunch 12:30 13:30
2 Synergi Plant RBI Extensive Demonstration 13:30 15:00 Demo & Hands- on
3 Likelihood theory 1 15:00 17:00 Thinning: calculation principles and inspection updating
Day 2
4 Likelihood theory 2 09:00 10:30 Other damage mechanisms
5 Consequence theory and exercises 10:30 12:30 Consequence theory
Lunch 12:30 13:30
6 Project Start up and Data Organization. 13:30 17:00 Inventory groups and Corr circuits
Day 3
7 RBI screening Introduction 09:00 11:00 Using Synergi Plant RBI and the screening tool
8 Detailed analysis - Data entry 11:00 12:30 Model creation and data entry issues.
Lunch 12:30 13:30
8 Detailed analysis - Data entry 13:30 17:00 Model creation and data entry issues.
Day 4
9 Establishing criteria and using the IP tool in Synergi Plant RBI 09:00 11:00 Inspection Planning using risk criteria
10 Plant inspection plans 11:00 12:00 Creating inspection plans from the RBI guidelines.
11 Reporting features / information output 12:00 12:30 Extracting information from Synergi Plant RBI
Lunch 12:30 13:30
12 Other Features 13:30 16:00 Synergi Plant RBI other features

DNV GL © 2014

INTRODUCTIONS
(Name,
Organisation,
Background
Type of work,
Why interested in RBI)

DNV GL © 2014 SAFER, SMARTER, GREENER

2
 15/06/2016

RBI TRAINING COURSE

RBI History

Probabilistic risk analysis techniques

– Started in the nuclear industry (1970s)
Quantitative risk assessment (QRA) in the Process Industries
– Canvey Island and the Rijnmond Report (1980s)
Software tools for QRA
– Eg DNV-Technica develops SAFETI and PHAST risk assessment tools (1980s)
ASME RBI principles overview document in 1991
API develops Risk Based Inspection Methodology (mid-1990’s)
– DNV main API sub-contractror
– API Base Resource Document 581 (2000)
– API RBI software
– API RP 580 (2002)

DNV GL © 2014

RBI History

DNV develops Synergi Plant RBI Onshore 1997-now

– Some Reasons:
– Need for a RBI software for all onshore installations
– API 581 focuses on refineries
– Improved consequence calculations with PHAST link
– Enhancements in likelihood calculation
– Synergi RBI uses equations for limit state implementation
– Need for a robust software architecture & professional software development
and maintenance
– Synergi Plant RBI Onshore is consistent with the API 580 RBI standard
– Synergi Plant RBI Onshore basically follows API 581
API RP 581 Update (2008)
API RP 580 Update (2009)
API RP 581 3rd edition (2016)- RBI Onshore 5.3.2 consistent with 3rd edition

DNV GL © 2014

3
 15/06/2016

RBI TRAINING COURSE

API Inspection and FFS Standards

Existing RBI & FFS documents

API RBI
750 API RP
580
API - BRD P 581
RISK BASED RBI
API INSPECTION API RP
510 581

MPC
API FITNESS FOR FFS
570 SERVICE API RP
579

ASME
API
653

Working Research & reference New

Documents Documents Documents

DNV GL © 2014

PROCESS CORROSION
- Continuously degrading integrity

DNV GL © 2014 SAFER, SMARTER, GREENER

4
 15/06/2016

RBI TRAINING COURSE

Corrosion Principles

Corrosion rate is measured as weight loss per unit area and is expressed in mils
per year (mpy) or mm/y.
Corrosion Rates can be affected by:
– Passivity forming protective surface films (including corrosion inhibitors, paints
and coatings)
– Oxygen content
– Flow velocity/rates
– Temperature
– pH effects (Low and High)
– Contaminants/intermediates

DNV GL © 2014

Some Corrosives Found In The Process Industry

Water Deposits
Oxygen Hydrogen Chloride
Naphthenic Acid Sulfuric Acid
Polythionic Acid Hydrogen
Chlorides Phenols
Carbon Dioxide Dimer and Trimer acids
Ammonia Other
Cyanides

DNV GL © 2014

5
 15/06/2016

RBI TRAINING COURSE

Low Temperature Corrosion

Below 500°F (<260°C)

Occurrences
– Inorganic compounds such as water, hydrogen sulphide, hydrogen chloride,
sulphuric acid, salts, etc.
– Presence of water (even in very small amounts)
– Electrolyte in hydrocarbon stream
– Hydrocarbons in water streams creating acidic conditions
– Solids .. Under deposit
– Organic acids
– Vapour Streams at water condensation points
Obeys electrochemical laws
Stable films can reduce or prevent corrosion

DNV GL © 2014

Low Temperature Corrosion

From Process chemicals

From Process contaminants
Not caused by clean hydrocarbons
Caused by inorganic compounds such as water, hydrogen sulphide, hydrogen
chloride, sulphuric acid, salts, etc.

DNV GL © 2014

6
 15/06/2016

RBI TRAINING COURSE

High Temperature Corrosion

Above 500°F (>260°C)

No water present
Result of a reaction between metal and process ions (such as oxygen O-, sulphur
S, etc.)

DNV GL © 2014

High Temperature Corrosion

Important due to serious consequences

High temperatures usually involve high pressures.
Dependent on the nature of the scale formed
– General thinning
– Localized thinning (pitting)
– Inter-granular attack
– Mixed phase flow
Metallurgical changes

DNV GL © 2014

7
 15/06/2016

RBI TRAINING COURSE

Situations Leading To Deterioration

Normal operation, upset, startup /shutdown conditions

Material/Environment condition interactions
Many combinations of corrosive process streams and temperature/pressure
conditions.
In the absence of corrosion, mechanical and metallurgical deterioration can occur.
Weather effects ….

DNV GL © 2014

Forms Of The Damage

General loss due to general or localized corrosion

Pitting attack
Stress Corrosion Cracking (SCC)
Metallurgical Changes
Mechanical damage
High Temperature Hydrogen Attack (HTHA)

Damage types occur with specific combinations of materials and environmental/

operating conditions

DNV GL © 2014

8
 15/06/2016

RBI TRAINING COURSE

Stress Corrosion Cracking Detection

SOHIC in soft base

metal.
Stress-Oriented
Hydrogen Induced
Cracking

In contrast to general
corrosion, SCC is very
hard to detect visually
even when it has
progressed to an
extreme condition.

DNV GL © 2014

Types of Stress Corrosion Cracking

Chloride stress corrosion cracking (Cl-)

Nitrates
Caustic stress cracking (NaOH)
Polythionic acid stress corrosion cracking
Ammonia stress corrosion cracking (NH4)
Hydrogen effects (in steel)
Sulfide stress corrosion cracking SSC, hydrogen induced cracking HIC, stress
oriented hydrogen induced cracking SOHIC
Hydrogen cyanide HCN
Others

DNV GL © 2014

9
 15/06/2016

RBI TRAINING COURSE

High Temperature Hydrogen Attack (HTHA)

Carbon and low alloys steels exposed to hydrogen above 430°F (221°C)
Hydrogen Partial pressure above 50 psi (>4 bar)
Dissociation of molecular hydrogen to atomic hydrogen
H2 -> 2 H+
Atomic hydrogen permeation into the steel
Reaction of atomic hydrogen with carbon in steel
Formation of methane at discontinuities
API 941 recommended for new installation

DNV GL © 2014

High Temperature Hydrogen Attack

Longitudinal Weld
Magnification: 500x Etch: 2% Nital
DNV GL © 2014

10
 15/06/2016

RBI TRAINING COURSE

Metallurgical And Environmental Failures

Grain growth Temper embrittlement

Graphitization Liquid metal embrittlement
Hardening Carburization
Sensitization Metal dusting
Sigma phase Decarburization
885 F embrittlement Selective leaching

DNV GL © 2014

Mechanical Failures

Incorrect or defective materials Over pressurization

Mechanical fatigue Brittle fracture
Corrosion fatigue Creep
Cavitation damage Stress rupture
Mechanical damage Thermal shock
Overloading Thermal fatigue

DNV GL © 2014

11
 15/06/2016

RBI TRAINING COURSE

Conclusions

There are many causes of equipment failures in the process industry.

Many are common and well documented.
Other, less common deterioration mechanisms are not well documented.
Deterioration is the result of metal and environment/ operating conditions
combinations.
These combinations vary somewhat in different process units.
Detection and characterization of the different forms is a challenging and critical
activity.

DNV GL © 2014

Tools exist to assist to assess the

severity of corrosion or determine the
appropriate materials of construction
For Example:

DNV GL © 2014 SAFER, SMARTER, GREENER

12
 15/06/2016

RBI TRAINING COURSE

NaOH Chart

DNV GL © 2014

These Tools Are Generally Used By

Experienced Corrosion Engineers.

They can also be implemented in

software as corrosion evaluation
supplements

DNV GL © 2014 SAFER, SMARTER, GREENER

13
 15/06/2016

RBI TRAINING COURSE

Determining Equipment Integrity

Requires information about the level of degradation:

– Monitoring (Fluid corrosivity) and

– Inspection (Wall condition)

DNV GL © 2014

“MONITORING” POSSIBILITIES

Monitoring
– Fluid Composition/Quality
– Pressure, Temperature, pH
– Contaminants when relevant
– Fluid corrosivity
– Corrosion probes (e.g. Weight loss, electrical resistance, linear polarization)
– Function of protective systems e.g. inhibitor injection

Inspection: Pressure boundary condition checks, e.g.

– Visual examination
– Thickness measurements
– Other checks

DNV GL © 2014

14
 15/06/2016

RBI TRAINING COURSE

Non Destructive Examination

- Inspection

DNV GL © 2014 SAFER, SMARTER, GREENER

NDE Methods
American Society for Nondestructive Testing (ASNT)

Acoustic Emission Testing (AE) Volumetric

Eddy Current Testing (ET) Surface/ Volumetric
Infrared/Thermal Testing (IR) Surface
Leak Testing (LT)
Magnetic Particle Testing (MPT) Surface
Neutron Radiographic Testing (NR) Volumetric
Penetrant Testing (PT) Surface
Radiographic Testing (RT) Volumetric
Ultrasonic Testing (UT) Volumetric
Visual Testing (VT) Surface
Magnetic Flux Leakage (MFL)

DNV GL © 2014

15
 15/06/2016

RBI TRAINING COURSE

Selecting Inspection method. Factors to consider

Type of defect
– General metal loss
– Localized metal loss
– Pitting
– Cracks
– Metallurgical changes
Location of defect
– On the outside wall of an item
– The inside wall
– Within the body of the wall
– Associated with a weld

DNV GL © 2014

Selecting Inspection method. Factors to consider:

Material of construction
– Magnetic
– Non magnetic
– Operating at high temperatures
– Insulated
Equipment geometry:
– May be hard to access
– May require extensive activity e.g. scaffolding, entry preparations, to perform
the inspection
Many considerations when determining how to inspect.
Also, should justify the need for inspection!

DNV GL © 2014

16
 15/06/2016

RBI TRAINING COURSE

Risk Based Inspection

DNV GL © 2014 SAFER, SMARTER, GREENER

Scope of RBI - Equipment Types

Pressure Vessels—All pressure containing components.

Process Piping—Pipe and piping components.
Storage Tanks—Atmospheric and pressurized.
Rotating Equipment—Only pressure containing components.
Boilers and Heaters—Pressurized components.
Heat exchangers (shells, floating heads, channels, and bundles).
Pressure-relief devices.

DNV GL © 2014

17
 15/06/2016

RBI TRAINING COURSE

What Constitutes an Undesirable Event In RBI?

Failure is defined as a leak of the equipment contents to the atmosphere; “breach

of containment” or LOPC
– Heat exchanger failures are channel or shell leaks.
– Pump failures are due to seal leaks and adjacent piping fatigue cracking.

DNV GL © 2014

RBI - Detailed Analysis

Components in the calculation of the risk

Consequence
Risk = Likelihood of Failure X of Failure
MF x Fp x Fm x Fu x GFF x DF Fdomino x CoF

Injury
Abbreviations:
: Age
Equip. Repair
DF: Damage
Factor Damage Business Int.
GFF: Generic Failure Type/Rate
Frequency Environ. Effects
Fi : Process, Mechanical
& Universal Factor
Fdomino:Domino Eff.Factor Inspection
MF: Management Factor Effectiveness

RBI_Key_Concepts.vsd

DNV GL © 2014

18
 15/06/2016

RBI TRAINING COURSE

Synergi RBI Damage Mechanisms/ Probability of failure (PoF)

Damage
Mechanisms

Stress
Internal External Brittle Fatigue HTHA Lining Creep
Corrosion PRVs
Thinning Cracking Damage Fracture

• HCl • Caustic General

. • Amine
• HT Sulfide
CUI
& Nap. Acid • SSC
• HT H 2S/H2 • HIC/SOHIC Cl SCC
• H 2SO4 • Carbonate
• HF • PTA
• Sour Water • ClSCC
• Amine • HSC-HF
• HT • HIC/SOHIC-HF
Oxidation
• CO2
Corrosion

DNV GL © 2014

CALCULATING THE FAILURE FREQUENCY

MANUAL ACTIVITY Damage factor Calculation

Estimate the likely
damage state / Determine the Likelihood of being in one
severity of the different possible damage states:
Consider data source Damage states

1 No worse than predicted X%

2 Up to 2x worse than predicted Y%
Assess the 3 Up to 4x worse than predicted Z%

inspection history
(Effectiveness) Calculate the failure frequency for each
Inspection Effectiveness
state using the relevant limit state
equation
Failures only occur when
the rate of degradation is Calculate the weighted failure frequency
higher than expected. for the item based on the Likelihood of
being in the different states.
Steps in Bayes_LoF

DNV GL © 2014

19
 15/06/2016

RBI TRAINING COURSE

Accident Consequence Simulation using PHAST Models

DISPERSION

Possible outcomes:
Toxicity fatalities
Environmental

DNV GL © 2014

Accident Consequence Simulation Using PHAST Models

IGNITION

Possible Flammable outcomes: Possible Hazardous effects

Fireballs/BLEVEs Radiation levels
Overpressure levels
Jet fires
Flash fire zones
Pool fires
Flash fires Consequences
Explosions Equipment Damage
Flammable Fatalities

DNV GL © 2014

20
 15/06/2016

RBI TRAINING COURSE

Amount of Effort - RBI vs QRA

Likelihood Consequence

QRA*

RBI**

* Quantitative Risk Assessment ** Risk Based Inspection

DNV GL © 2014

Input Data For A Quantitative RBI Assessment

The main input data collected

Item Design Data Operating Data Damage mechanisms Inspection data
OD Tnom Matl Ins Press Temp Fluid Temp. Press Fluid Mechanism Severity/rate Done? Result?
A Thinning,
SCC,
Furnace,
HTHA,..
B
C

Identify For some damage Is it What do we What has been

all items mechanisms, e.g. SCC, operating as expect to find looked for and
brittle fracture, fatigue, intended? and what at what has been
other data may be what severity? found
needed e.g. PWHT,
Charpy test temp.

DNV GL © 2014

21
 15/06/2016

RBI TRAINING COURSE

RBI Results?

Calculation of the risk with a lookahead: Inspection Plan

Item Type From To Damage GFF DF LoF CoF Risk Insp. Insp. New
no. Mechanism Type Date DF

1 Pipe Thinning 3000

2 Vessel CUI 100
3 Fin Fan Erosion 0.5

What - Why -
(Damage mech. &
(Risk priority)
factor) When -
(Basis Inspection planning
Where / How - targets.)
(Item - Effectiveness - Material - Mechanism)

DNV GL © 2014

The Presentation Of Risk

5 Medium-High Risk High Risk

4 Med. High Risk

Likelihood
Category

2 Medium Risk

1
Low Risk

A B C D E
Consequence Category

DNV GL © 2014

22
 15/06/2016

RBI TRAINING COURSE

How Will This Picture Change With Time?

Likelihood of Failure

A B C D E
Consequence of Failure
DNV GL © 2014

Risk Increase Over Time

Likelihood of failure
will increase over
time because of time-
dependent material
Likelihood of Failure

degradation

A B C D E
Consequence of Failure
DNV GL © 2014

23
 15/06/2016

RBI TRAINING COURSE

What is the effect of Inspection ?

Likelihood of Failure

A B C D E
Consequence of Failure
DNV GL © 2014

Inspection Program Options for Influencing Risk

Change inspection frequencies (when)

Change inspection scope / thoroughness (what)
Change inspection tools / techniques (how)

DNV GL © 2014

24
 15/06/2016

RBI TRAINING COURSE

RBI - Applications

Risk-prioritized Turnaround planning

– High safety/reliability impact = more attention (in order to lower risk)
– Less impact safety/reliability = less attention (in order to lower costs)
– Result:
– Lower equipment life cycle costs
– Fewer incidents / outages
– Fewer unnecessary inspections
– Higher reliability
May also assess the impact of delaying a turnaround/ shut down
What if studies e.g.
– Assess the impact of process changes.
– Assess the impact of a different feed

DNV GL © 2014

Can RBI Help To Prevent All Releases?

DNV GL © 2014 SAFER, SMARTER, GREENER

25
 15/06/2016

RBI TRAINING COURSE

Where Inspection Can Help

About half of the Mechanical

containment Failure
losses in a 43%
typical Process
petrochemical 11% Upset
process plant 1% Sabotage/Arson
can be Natural Hazard
influenced by 5%
Unknown
inspection 14%
Operational
activities
Error
21%

Design Error
5%

DNV GL © 2014 Source: Large Property Damage Losses in the HC-Chemical Industries - A thirty year review, 17th edition, J&H Marsh& McLennan.

Presentation Topics

General Introduction
The benefits of RBI
What RBI is
How RBI fits within existing plant systems
Implementing RBI
Some case studies

DNV GL © 2014

26
 15/06/2016

RBI TRAINING COURSE

Typical RBI implementation

Define scope of RBI Study
Set up RBI team and train
Collect Data
Identify inventory groups (For consequences)
Identify Corrosion circuits
Perform Screening Analysis
Select high risk equipment items for Detailed Analysis
Perform detailed RBI analysis
– Consequence input data-Likelihood input data
– Run risk assessment & Review the results
– Develop action criteria
– Discuss Synergi Plant RBI proposed inspection guidelines and run final
Translate into an actual inspection plan with schedule
Implement plan-perform inspections
Update the model with latest inspections
DNV GL © 2014

Risk Target and Inspection Planning

Risk / DF

Inspection
Risk / Damage Factor (DF)

Target

Fairly Effective

Highly Effective
Predicted Risk
Increase
Time to next
inspection

Now
1st 2nd
Time
Turnaround Turnaround

DNV GL © 2014

27
 15/06/2016

RBI TRAINING COURSE

Qualitative vs Quantitative - COST COMPARISON

For repeat analyses the quantitative approach is far more efficient.

The benefits multiply with time

Proportion of the time spent on activity:

Est. total Accum.

Method Activity Hours on "Value"
hours Hours
Data Coll. Analysis Insp. plan insp plan
Initial Analysis
Qual. 310 10% 40% 50% 155 na 40

Quant 500 60% 10% 30% 150 na 100

Second time around:
Qual. 310 10% 40% 50% 155 620 40

Quant 200 15% 15% 70% 140 700 100

DNV GL © 2014

ADVANTAGES OF THE QUANTITATIVE APPROACH

Not simply opinion based-easily reproducible

Accuracy-Time model
– The results of qualitative and semi quantitative studies are frozen in time. In
reality the risk will change as the equipment ages and as new data is available
from inspection. The quantitative method incorporates this.
What if studies, e.g.:
– New campaigns in swing plants
– If the study had been done qualitatively or semi quantitatively, the effort would
be much higher
– i.e. It is more efficient and powerful to use an analytical approach

DNV GL © 2014

28
 15/06/2016

RBI TRAINING COURSE

Issue:
Should we change our feed to a cheaper but more
corrosive alternative?
What does this mean for our risks and inspection
requirements?

EXAMPLE STUDY 1

DNV GL © 2014 SAFER, SMARTER, GREENER

Example Study 1

Maximum Tolerable Risk Unacceptable Risk

Corrosive Conditions
Standard
Operating
Risk Conditions

Tolerable Risk Changed Inspection

Frequency
Inspection Interval

DNV GL © 2014

29
 15/06/2016

RBI TRAINING COURSE

Example Study 1

Financial Risk Exposure

($ per year per equipment item)
Financial Risk after Inspection

$65,000

$55,000

$45,000 $46,846

$35,000
$34,793
$25,000
$26,421
$15,000

0.1% 0.5% 0.8%

Corrosive in the feed

DNV GL © 2014

Example Study 1

Cost of Inspection
Cost of Inspection

$350,000
$300,000
$250,000
$200,000
$150,000
$100,000
$50,000
$0
0.1% 0.5% 0.8%

% Corrosive in Process Feed

DNV GL © 2014

30
 15/06/2016

RBI TRAINING COURSE

Example Study 1

The study gave the facility the information on:

– The increased risk exposure
– The increased cost of inspection
They compared this with the cost benefits of the cheaper feed and made their
decision.

DNV GL © 2014

Example Study 2

Current Inspection Costs

Current Maintenance Costs
$1,400,000
Total Current Costs
RBI Inspection Costs
$1,200,000 RBI Maintenance Costs
RBI Total
$1,000,000 Total Savings

$800,000

$600,000

$400,000

$200,000

$0
Unit 30 Unit 33 Unit 34 Unit 48 Unit 51
-$200,000

-$400,000

DNV GL © 2014

31
 15/06/2016

RBI TRAINING COURSE

Example Study 2
COST BENEFIT ANALYSIS Results for all Units

$3,000,000

$2,500,000

$2,000,000

$1,500,000

$1,000,000

$500,000 Current
RBI
$0
Savings

DNV GL © 2014

Example Study 3

Cost effective decision making for an older refinery with a limited inspection
history.

DNV GL © 2014

32
 15/06/2016

RBI TRAINING COURSE

Using The Financial Risk Values

Total Risk vs. Risk Rank

Refinery Process Unit, Top 10% Risk Items
$1,400,000

$1,200,000

$1,000,000
Total Risk =
Risk,$/yr

$800,000

$600,000
$11,500,000/year
$400,000

$200,000

$0
0 10 20 30 40 50

Risk Rank

DNV GL © 2014

Assess The Cost Benefits Of Inspection

Total R is k vs . R is k R ank
R e fine ry Proce s s Unit, Top 10% R is k Ite ms ,
$ 1 ,2 0 0 ,0 0 0
Same Ite ms , Each with 1 M ore Ins pe ction
$ 1 ,0 0 0 ,0 0 0
Total Risk = $4,100,000/yr,
$ 8 0 0 ,0 0 0
Savings = $7,400,000/yr
Risk, $/yr

$ 6 0 0 ,0 0 0 Cost = $250,000 (mostly piping,

approximately $5,000 avg. insp.
$ 4 0 0 ,0 0 0
cost)
$ 2 0 0 ,0 0 0

$0
0 10 20 30 40 50

Ris k Rank

DNV GL © 2014

33
 15/06/2016

RBI TRAINING COURSE

The Risk of the Lowest 10% Items

Total Risk vs. Risk Rank

Refinery Process Unit, Bottom 10% Risk Items
$1,600
$1,400

$1,200 Total Risk =

$1,000 $12,000/yr
Risk, $/yr

$800
$600

$400
$200
$0
0 10 20 30 40 50

Risk Rank

DNV GL © 2014

The Inspection Benefits Here

To ta l R is k v s . R is k R a nk
R e fine ry Pro ce s s Unit, B o tto m 10 % R is k Ite ms ,
$ 1 ,2 0 0
S ame Ite ms , Each with 1 M ore Ins pe ctio n
$ 1 ,0 0 0
Total Risk = $4,300/yr,
$800
Risk, $/yr

$600
Savings = $7,700/yr

$400 Cost = $250,000 (mostly

piping, approximately $5,000
$200
avg. insp. cost)
$0
0 10 20 30 40 50

R is k R ank

DNV GL © 2014

34
 15/06/2016

RBI TRAINING COURSE

[email protected]

www.dnvgl.com

SAFER, SMARTER, GREENER

163 DNV GL © 2014

35