2016 - d2 03 Borger Paper
2016 - d2 03 Borger Paper
Abstract
DIN 18009 Part 1 describes the fundamentals of the system for the selection of design
fire scenarios. The working steps can only be regarded as indications for a performance-
based design. This framework is supposed to ensure an adequate and uniform application
of the variety of internationally available knowledge and methodologies.
In 2008, the DIN working committee NA 005-52-21 AA started its work to standardize
fire safety engineering methods in Germany. The second sub document DIN 18009 Part 3
shall cover the design fire scenarios and design fires. Some ideas related to methodology,
requirements and scenario-based design are presented in this article.
INTRODUCTION
In Germany, performance-based design methods are not explicitly embedded in national build-
ing codes. Accordingly, the major knowledge, guidance as well as material and product per-
formance is closely linked to the prescriptive design approach. This approach has provided a
substantial improvement of fire safety, resulting in a quite safe situation within the built envi-
ronment at the present time. However, the increasing individuality and complexity of modern
buildings or questions related to the conversion of existing buildings may conflict with the ap-
plicability or fulfillment of those deemed-to-satisfy requirements. Thus, the performance-based
fire safety design has established in Germany as well.
Consequently, international standards are implicitly incorporated in the German day-to-day
business of FSE. The current work on the DIN 18009 framework is supposed to ensure an
adequate and uniform application of the variety of internationally available knowledge and
methodologies [13].
1
industrial buildings. Basic principles for fire safety engineering are solely published in sub-
statutory reports, guidelines and specifications.
The goal of the German Fire Protection Association (GFPA) (Vereinigung zur Förderung des
Deutschen Brandschutzes e. V. – vfdb) Section 4 “Fire Protection Engineering” is to upgrade
the modern fire protection engineering methods developed in the last years. Further-on it is the
aim to make this upgrade available for daily applications in terms of of a guideline [9]. In the
scope of fire protection concepts, this guideline is intended to contribute to harmonize disjointed
approaches and assumptions and to avoid erroneous measures in the application of engineering
methods.
DIN 18009 Part 1 is intended as basis document to standardize the methodology in Fire Safety
Engineering. It describes all characteristic steps and necessary terms and definitions related to
the design process. In this respect, it is intended to serve as a guideline for both the design
and the inspection process. In accordance with the basic principles of DIN 18009 Part 1, it is
explicitly allowed to use other national or international standards in order to supplement the
framework.
In principal, DIN 18009 Part 1 comprises the following engineering methodologies: performance-
based, argumentative and experimental; the emphasis is placed on the performance-based line.
Irrespective of the question and the chosen method, the fundamental proof is based on the
question if the system’s resistance is greater than the minimal required resistance for a specific
influence. To encourage this concept, guidelines for the identification of fire safety goals, per-
formance criteria and acceptance criteria are provided. Further on, the framework regulates a
classification of scenarios as well as the setup and selection of the latter, see Figure 1. Finally,
a concept for the inclusion of safety margins and guidelines for the documentation are provided
to the practitioners [13].
The first sub document DIN 18009 Part 2 covers evacuation modeling and the life safety as-
sessment during fire and is currently in a first draft state. A special focus of the draft is the
identification of representative scenarios and the related translation into parameter samples.
Regarding the established model classes to describe pedestrian dynamics, the methodology is
designated to cope with different extends of data input and output. A variety of recommenda-
tions for data analyses, visualization and documentation are addressed. Selected insights related
to methodology, requirements and scenario-based design are presented in [10].
2
DIN 18009 Part 3
The second sub document DIN 18009 Part 3 covers the design fire scenarios and design fires.
A special focus will be the identification of representative scenarios, the design fires and the
selection of the fire model. The definition of explicit design fires like [11, 12] are in discussion.
Some ideas related to methodology, requirements and scenario-based design are presented in
the next section.
Start
1.
Identification of fire scenarios based on building utilization
2.
Definition of critical sites for the source of fire
5.
Assessment of the estimated fire losses
6.
Assessment of fire risk
7.
Selection and documentation of the relevant scenarios
End
3
SELECTION OF DESIGN FIRE SCENARIOS
Performance criteria
A performance based fire protection design serves to ensure a safety level in accordance with
building regulations. The evidence should be provided by comparing the defined performance
criteria on the resistance side (R) with the respective calculation results on the action side (A).
The influencing variables must be selected based on sufficiently conservative assumptions or
have to be varied within parameter studies. In order to take account of uncertainties on the part
of the individual parameters, the calculation of corresponding partial safety coefficients γA and
γR , is generally carried out as part of a safety concept [3].
R · γR ≥ A · γA
Table 1 provides an example of how quantitative and qualitative requirements can be derived
from the respective fire safety goals.
4
Table 1: Fire safety goals and performance criteria [9]
Fire safety Functional requirement and qual- Performance criteria for quanti-
goal itative verification tative verification
Prevent the Restriction of combustible building materials. Verification by:
fire from
starting - Proof of usability (abZ, abP) - Norm specifications
- Test - Test and authorization criteria
Prevent the Limitation of the fire effects to one utilization by:
spread of fire
and smoke - Fulfillment of material demands - Test criteria for integrity and/or
on partitioning structural elements smoke tightness
- Proof of fire effects - Minimum distance to adjacent
building
- Norm specifications for room
closing components
- Max temperature or heat radiation
Within a qualitative analysis, the design and utilization of the building, as well as the scope of
technical fire protection systems shall be recorded. In particular, from the type of use, potential
sources of ignition can be derived from the furniture or the stored substances. Furthermore it is
necessary to determine which fire products or which combustion heat could be released by the
substances in the event of a fire, as well as the resulting risk to people, property values or the
load-bearing structure.
The following events must be considered by means of an individual analysis or taking into
account fire statistics:
For individual fire scenarios, different locations are generally considered to be the starting point
of a fire with a comparable probability. From this, those are to be selected for further consid-
eration, which are particularly critical regarding the fire safety goal to be achieved. In order to
verify that a successful evacuation of the building is possible, for example, fires in the area of
the escape routes or fires which make escape routes impassable (e.g. fires in atria) can be con-
sidered. In terms of ensuring effective fire-fighting, fire sites are particularly relevant, which are
characterized by an extraordinary speed of propagation and which result in a high thermal load
on the load-bearing structure. Overall, fires within and outside the building must be considered.
The influences on the course of the fire must be derived and quantified from the structural and
geometrical as well as technical fire protection conditions. By the location and arrangement of
room-closing components, burning areas can be defined, to which the fire event can be reduced.
Furthermore, an assessment can be made as to how far a fire impact on adjacent areas can
be expected (Traveling Fire). Based on the opening surfaces (windows, doors) including their
respective opening criteria, information about the ventilation can be obtained.
6
Technical fire protection measures can be used to limit unrestricted fire spreading directly (au-
tomatic fire-extinguishing systems) and indirectly through early fire detection and alarming of
the users or the fire brigade. The influence on the course of the fire must be determined taking
into account plant-related failure probabilities.
A fire event can either be discovered and reported by a person or automatically by appropriate
detection systems. Depending on the physical and mental condition of the building users, the
respective reaction features can be estimated for such an event. From the use of the building,
conclusions can usually be drawn to the vigilance, the ability to act independently and the
familiarity with the building of the users.
The following factors are particularly relevant to the influence on the course of the fire:
Since the reactions of the building users are extremely difficult to estimate or are subject to
large scatter, the influences must be taken into account by the application of appropriate safety
factors. Alternatively, conceivable scenarios can be covered by parameter studies.
On the basis of the previously defined fire scenarios, the respective consequences should be
estimated taking into account the respective fire safety goals. This can be done, for example,
by quantifying personal injury or economic damage, such as damage to property or operational
failures.
By combining the frequency with the consequences, a fire risk can be determined for the indi-
vidual fire scenarios. If the fire scenarios can not be assigned to different frequencies, all fire
scenarios found should be assumed to be equally probable.
For a further consideration, the fire scenarios with the highest risk should be selected. Further-
more, it may be necessary to analyze individual unlikely scenarios in a deterministic way.
The identification and definition of design fire scenarios operates in mutual interaction with
the subsequent fire simulation, so that the entire procedure is to be understood as an iterative
process. Based on the calculation results of the fire simulation it may be necessary to adjust
input parameters of the scenario. For example, the probable triggering time of an automatic fire
extinguishing system can only be determined in relation to the calculated fire temperatures.
7
Scenario-based design
The selection of one or more relevant design fire scenarios should be representative of a group of
potential scenarios. For the selection of the scenarios, an assessment period is first established,
which generally corresponds to the service life of a building. All possible scenarios must be
recorded over this period. In principal, the totality of possible fire scenarios are grouped into
relevant and non-relevant scenarios. By means of risk, non-relevant scenarios can be differen-
tiated in bagatelle and worst-case scenarios. The risk of a particular scenario is defined as the
product of the probability of occurrence and the amount of damage.
Additionally, a trial design may also result in unacceptable scenarios with high risk potentials,
which categorically requires a redesign. Relevant scenarios are the centerpiece of the design
process. They comprise a set of significant scenarios which are supposed to be represented by
design scenarios. The outlined classification is illustrated in Figure 2.
An assessment of the scenarios can be carried out in the course of a risk analysis with determin-
istic methods, for example by expert knowledge or by probabilistic methods.
8
Fire modeling
Design fires as a basis for fire simulations are described by means of a time-dependent heat
release. In addition to normative fire curves, it is also possible to refer to experimentally deter-
mined fire sequences or engineering procedures. Standardized procedures for calculating heat
release curves regarding different fire safety goals (see DIN 18232-2 [5], DIN EN 1991-1-2 [6]
or DIN EN 1991-1-2/NA [7]) are characterized in particular by the following input variables:
• Fire load
Depending on the fire safety goals to be achieved or the functional requirements derived there-
from, it may be necessary to define a detailed combustion reaction in order to take account of
the formation of toxic fire products or soot.
The approach of a fire curve with a square increase in heat release, a constant fully-developed
fire stage and a linear decay stage is suitable for many fire protection verifications. In this case,
a radial fire propagation is assumed, which is limited by the maximum fire load, the ventilation
conditions or extinguishing measures. After 70% of the calculated fire load has been burned,
the heat release decreases linearly to zero.
9
Workflow
The DIN 18009 Part 1 describes the basic procedure for the selection of design fire scenarios
in seven working steps (see Figure 1). These are explained and supplemented by the following
flow chart in Figure 4.
Potential sources
Identification of building of ignition Identification of fire scenarios by :
Definition of fire safety goals
use, design and technical -Common causes of fire
1 and performance criteria
fire protection -Serious economic and personal damage
Hazards
Selection of scenarios
2 Identification of
with worst impacts
possible fire locations
on fire safety goals
no yes
no yes
3
/ Open doors and windows?
4
no yes no yes
Extinguishing systems?
Estimation of consequences:
4 -Personal damage
-Material damage
10
APPLICATION EXAMPLE
Description of building
The example shows a three-storey building with differently used areas. The ground floor in-
cludes a lobby, a restaurant and offices. The floors are connected by an atrium and open stairs.
All floors above the ground floor have a similar layout and include offices, meeting rooms and
two independent stairwells. The floor plans can be found in Figure 5.
Figure 5: Floor plans of the application example consisting of the ground floor and the two
overlying floors.
The building is built in solid construction, the individual units are separated from the atrium
by glass elements. The roof structure, which at the same time forms a bracing element of the
overall structure, is designed as an unprotected steel frame construction.
Since some of the stated objectives can not be achieved by fulfilling prescriptive requirements,
appropriate compensation measures must be taken or evidence must be provided to ensure an
equivalent level of safety.
11
In order to enable effective fire fighting, a sufficient stability of the load-bearing structure in
case of fire must be ensured. The required fire resistance of at least 90 minutes according
to building regulations can not be guaranteed by the unprotected steel construction without
further evidence. The evidence of compliance with the fire safety goal shall be provided by an
computational analysis based on a standardized natural fire according to DIN EN 1991-1-2/NA
[7].
Due to the utilization, for example, defective electrical equipment such as computers or kitchen
equipment as well as the misbehavior of persons can be identified as possible causes of fire. The
individual office units as well as the restaurant or the lobby have to be considered as potential
locations for the development of a fire. With regard to the fire safety goal to be fulfilled, in
particular, scenarios are to be considered which are expected to result in a high thermal load on
the roof structure. Taking into account the higher fire load as well as the location, the closer
examination is reduced to a fire event in one of the office rooms on the second floor.
Within the the second floor, a total of three areas of different sizes, which can be viewed as po-
tential fire areas, are formed by room closing components. Since the probability of occurrence
for a specific fire event can not be quantified in detail, it must be assumed to be identical for the
three office units because of the comparable utilization. The risk analysis is thus limited to the
selection of scenarios with the greatest possible impact.
Taking into account the maximum possible fire area (Af = 379 m2 ) as well as the ventilation
conditions, a full fire of the office unit 2 can be identified as the decisive design fire scenario
with regard to the maximum heat release as well as the fire duration.
Design fire
Inputs that are related to building utilization such as heat release rate, fire spread rate and fire
load density are calculated according DIN EN 1991-1-2 [6, 7]. The consideration of uncertain-
ties and failure probabilities of human and technical nature are covered by the safety concept
according to DIN EN 1991-1-2/NA [7] (see Table 2).
Table 2: Probabilities of occurrence and failure according to safety concept of DIN EN 1991-
1-2/NA [7]
Probability . . . p1 p2,1 p2,2 p3
. . . of occurrence for an initial fire in an utilization unit 6.200e − 03
. . . of the failure of firefighting measures by users 0.5
. . . of the failure of firefighting measures by the fire brigade 0.2
. . . of the failure of firefighting measures by an automatic 1.0
extinguishing system
12
The consequences for human life, as well as economic and environmental damage to be ex-
pected in the event of a failure of the load-bearing structure, can be estimated depending on the
utilization of the building. As a result, a required reliability and a permissible failure probabil-
ity can be determined. The permissible failure probability pf of the load-bearing structure in
relation to one year (office building) can therefore be defined as:
pf = 1.300e − 05
From this the conditional failure probability in case of fire pf,f i and the linked reliability index
βf i can be determined as follows:
pf 1.300e − 05
pf,f i = = = 2.097e − 02
pf i 6.200e − 04
The partial safety coefficients can be calculated with the variation coefficients Vq = 0.3 (fire
load density) and Vq = 0.2 (heat release rate). The sensitivity coefficient is assumed to be
α = 0.6 so that the following partial safety coefficients can be calculated:
γf i,Q̇ = 0.986
γf i,q = 0.981
The Eurocode specifies the characteristic values for the fire load density qf,k and the heat release
rate per unit area RHRf as 90% quantile. For office areas, the following values should be
assumed:
Figure 6 shows the change in heat release for the design fire as a function of time:
13
100
80
60
HRR [MW]
Flashover
40
20
0
0 10 20 30 40 50 60 70 80
Time [min]
I II III
14
CLOSING REMARKS
The contents of this article reflect the opinion of the authors and may not be misinterpreted as
a direct excerpt of the future standard DIN 18009-3. It may not be understood as the documen-
tation of a complete assessment process. It is rather supposed to record the methodological,
technical and regulative considerations made in the past months. Furthermore we aim for an
international exchange and discussion, which is warmly welcome. The presented approaches,
assumptions and conclusions may not be misinterpreted in any context of the prescriptive regu-
lations in Germany.
References
[1] Fachkommission Bauaufsicht der Bauministerkonfererenz (ARGEBAU). Musterbauord-
nung (German Model Building Code) – MBO. 2016.
[2] Fachkommission Bauaufsicht der Bauministerkonfererenz (ARGEBAU). Muster-Industriebau-
Richtlinie (German Model Industrial Building Code) – MIndBauRL. 2014.
[3] Heinrich Bub et al. Eine Auslegungssystematik für den baulichen Brandschutz. Ed. by
Schmidt. Brandschutz im Bauwesen 4. Berlin, 1983.
[4] DIN - Deutsches Institut für Normung. DIN 18009-1:2016-09 Fire safety engineering –
Part 1: Basic principles and codes of practice. Standard. 2016.
[5] DIN - Deutsches Institut für Normung. DIN 18232-2 2007-11: Smoke and heat control
systems – Part 2: Natural smoke and heat exhaust ventilators; design, requirements and
installation. Standard. 2007.
[6] DIN - Deutsches Institut für Normung. DIN EN 1991-1-2 2010-12 Eurocode 1: Actions
on structures - Part 1-2: General actions - Actions on structures exposed to fire. Standard.
2010.
[7] DIN - Deutsches Institut für Normung. DIN EN 1991-1-2/NA 2015-09 Eurocode 1: Na-
tional Annex – Nationally determinded parameters – Eurocode 1: Actions on structures
- Part 1-2: General actions - Actions on structures exposed to fire. Standard. 2015.
[8] DIN - Deutsches Institut für Normung. E DIN 18009-1:2015-04 Brandschutzingenieur-
wesen – Teil 1: Grundsätze und Regeln für die Anwendung. Standard. 2015.
[9] Dietmar Hosser. Leitfaden Ingenieurmethoden des Brandschutzes. Report vfdb TB 04-
01. Vereinigung zur Förderung des Deutschen Brandschutzes e.V., 2013.
[10] Gregor Jäger and Benjamin Schröder. “Evacuation and Life Safety Assessment in Ger-
many”. In: Proceedings of 11th Conference on Performance-Based Codes and Fire Safety
Design Methods. Ed. by Piotr Tofilo. Warsaw: Society of Fire Protection Engineers, 2016,
pp. 37–50.
[11] Innovation Ministry of Business and New Zealand Employment Wellington. Verification
method: framework for fire safety design, C/VM2. 2013.
[12] National Fire Protection Association. NFPA 101: Life Safety Code. 2015.
[13] Jürgen Wiese. “Ingenieurmäßige Brandschuztnachweise nach DIN 18009 sowie interna-
tionalen Regelungen - Normative Unterstützung für Aufsteller und Prüfer”. In: Braun-
schweiger Brandschutz-Tage 2014 28. Fachtagung Brandschutz bei Sonderbauten, 16.
und 17. September 2014: Tagungsband. Braunschweig, Sept. 2014, pp. 95–110.
15