Journal of the Institute of Conservation

ISSN: 1945-5224 (Print) 1945-5232 (Online) Journal homepage: https://www.tandfonline.com/loi/rcon20

The QuiskScan—a quick risk scan to identify value

and hazards in a collection

Agnes W. Brokerhof & Anna E. Bülow

To cite this article: Agnes W. Brokerhof & Anna E. Bülow (2016) The QuiskScan—a quick risk
scan to identify value and hazards in a collection, Journal of the Institute of Conservation, 39:1,
18-28, DOI: 10.1080/19455224.2016.1152280

To link to this article: https://doi.org/10.1080/19455224.2016.1152280

Published online: 24 Mar 2016.

Submit your article to this journal

Article views: 1977

View Crossmark data

Citing articles: 4 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rcon20
Journal of the Institute of Conservation, 2016
Vol. 39, No. 1, 18 –28, http://dx.doi.org/10.1080/19455224.2016.1152280

Agnes W. Brokerhof and Anna E. Bülow

The QuiskScan—a quick risk scan to identify value

and hazards in a collection

Keywords
risk assessment; collection management; cultural value; vulnerability; preventive conserva-
tion; decision making

The challenge: getting a useful overview of risks in a short period of

time
The museum, archive and library world has had approximately two
decades of experience with risk assessment and risk management for col-
lections, and the majority of such organizations will have had a procedure
for managing corporate and security risks for much longer. From the per-
spective of collection preservation an integrated approach has been
applied in which safety and security are combined with preventive conser-
vation and facility management to minimize losses to the cultural assets of
that organization. Robert Waller’s Cultural Property Risk Assessment
Method (CPRAM) and Stefan Michalski’s ABC method are probably the
1 R.R. Waller, Cultural Property Risk best known approaches for this ‘collection risk management’.1 Both
Analysis Model: Development and Appli- methods are in line with the framework of the ISO 31000, and identify
cation to Preventive Conservation at the risks systematically using the 10 agents of deterioration: physical forces,
Canadian Museum of Nature, Göteborg
Studies in Conservation 13 (Göteborg: fire, water, thieves and vandals, pests, contaminants, light and UV/IR radi-
Göteborg Acta Universitatis Gothobur- ation, incorrect relative humidity, incorrect temperature and dissociation.2
gensis, 2003); S. Michalski and J-L. Ped- Over the years both methods have been taught and applied in a consider-
ersoli, Reducing Risks to Cultural able number of heritage institutions.3
Heritage Course Manual and Database These concepts have been embraced by the heritage profession. Waller
(Resources of the ICCROM-CCI-RCE
‘Reducing Risks to Heritage’ Course, and Michalski’s methods have had a fundamental impact on the paradigm
2011), http://www.reducingrisks.info shift in preservation thinking.4 Together with Jonathan Ashley-Smith’s
(accessed 9 September 2015). ground breaking work,5 they have laid the foundations for risk-based
2 ISO 31000, Risk Management—Principles decision making in conservation and for risk-based guidelines for preven-
and Guidelines (Geneva: International tive conservation such as those for lighting and managing environmental
Organization for Standardization, 2009), conditions for cultural collections.6 Yet the application of such methods
http://www.iso.org/iso/iso_catalogue/ continues to meet with some resistance. Despite courses and workshops,
catalogue_tc/catalogue_detail.htm?csnu
training and coaching, and the development of tools to support the
mber=43170 (accessed 2 September 2015);
S. Michalski, ‘A Systematic Approach to process, most organizations fear dedicating the resources required to
Preservation: Description and Integration carry out such a systematic and detailed risk assessment.7 In addition,
with Other Museum Activities’, in Pre- there can be a perception that such work is inefficient because at the end
prints of the Contributions to the Ottawa Con- of such a process the amount of seemingly unnecessary work executed
gress, 12–16 September 1994, Preventive
does not obviously contribute to the preservation of the collection. This is
Conservation: Practice, Theory and Research,
ed. A. Roy and P. Smith (London: The a significant threshold to overcome, notwithstanding that the outcomes
International Institute for Conservation and impact are generally appreciated by those that have gone through
of Historic and Artistic Works, 1994), 8– the process.
11; R.R. Waller, ‘Conservation Risk Assess- Such an inertia is made more acute where detailed and intensive
ment: A Strategy for Managing Resources
methods of risk analysis are set against the wider climate of budget cuts
for Preventive Conservation’, in Roy
and Smith, Preprints of the Contributions and staff reductions. Nowadays collection care planning and activities
to the Ottawa Congress, 12–6, require speed and cost-effectiveness, whereas currently available
http://www.museum-sos.org/docs/Wa methods are time-consuming as they follow the risk management process
llerOttawa1994.pdf (accessed 2 September from the first to the last step systematically. Risk identification is followed
2015).
by detailed analysis so that all risks can be evaluated and compared to
3 See for example, C. Antomarchi,
A. Brokerhof and J. Stevenson, ‘Reducing
(Received 26 November 2015; Accepted 5 February 2016)

# 2016 Icon, The Institute of Conservation

The QuiskScan—a quick risk scan to identify value and hazards in a collection 19

develop mitigation options for the relevant risks and the best choices can be Risks to Cultural Heritage: Analysis of a
decided upon. For many institutions, such a process is cost-prohibitive. Course Metamorphosis’, in ICOM–CC
However, organizations such as the British Museum (UK), the Netherlands 17th Triennial Conference Preprints, Mel-
Open Air Museum (Arnhem, NL) and museums in the city of Leiden (NL) bourne, 15–19 September 2014,
ed. J. Bridgland (Paris: International
have been willing to invest a limited amount of time in engaging with risk
Council of Museums, 2014), art. 0301, 8;
management either to get an overview of their risks or of risks in specific A.W. Brokerhof, T. Luger, B. Ankersmit,
situations, or to assess whether the correct mitigation measures were in F. Bergevoet, R. Schillemans,
place. These needs have triggered the development of a tool that quickly P. Schoutens, T. Muller, J. Kiers,
provides the first systematic insights into risks without going into in- G. Muething and R. Waller, ‘Risk Assess-
ment of Museum Amstelkring: Appli-
depth analysis of specific risks. Such a tool would not replace a full risk
cation to an Historic Building and its
assessment but would serve to generate a first integral overview on Collections and the Consequences for
which an organization may choose to build a detailed risk assessment. Preservation Management’, in Preprints
of the ICOM–CC 14th Triennial Meeting,
Different approaches to risk assessment The Hague, The Netherlands, 12–16 Septem-
ber 2005 (London: James & James, 2005),
Risk is usually defined as something negative such as ‘the possibility of
590–6; A.E. Bülow, ‘Risk Management as
loss’. It is a combination of the likelihood and the consequence of something a Strategic Driver for a Large Archive’,
happening that causes a loss. All methods for risk assessment seek to Collections: A Journal for Museums and
answer three questions: Archives Professionals 5, no. 9 (2009): 61–
72; A.E. Bülow, ‘Collection Management
1 What might happen? This is often thought of as a one-off event but can Using Preservation Risk Assessment’,
also be an ongoing process. Journal of the Institute of Conservation 33,
no. 1 (2010): 65–78; L.K. Elkin,
2 How likely is that? How often will the event happen or how fast will the D. Fenkart-Froeschl, E. Nunan and
process unfold? R. Waller, ‘A Database Tool for Collections
3 What will be the consequences? How bad will it be when the event or Risk Evaluation and Planning’, in ICOM–
process happens? CC 16th Triennial Meeting Preprints, Lisbon,
Portugal, 19–23 September,
with underlying questions about: ed. J. Bridgeland and C. Antomarchi
(CD_ROM, ICOM–CC and Almada Cri-
a Which hazards do we worry about? terio, 2011); M. McCubbin, A. Cannon,
C. Carter, D. Henry, H. Privett, N. Ladas,
b Are the assets exposed to those hazards? D. Leggett, R. Leveson, M. Raberts,
L. Stedman and R. Waller, ‘Improving
Both the ABC and CPRAM methodologies describe risks as scenarios Risk Assessment Methods in a Complex
with a cause – pathway –effect structure. The causes all act through one or Setting: Museum Victoria’s Collection
more of the agents of deterioration. However, the root cause may lie Risk Assessment’, in Bridgland, ICOM–
beyond the immediacy of those agents. For example, collections may get CC 17th Triennial Conference Preprints;
A. Paolini, A. Vafadari, G. Cesaro,
wet in a flood which is a risk which, ultimately, can be argued to have
M. Santana Quintero, K. Van Balen,
been caused by human activity causing climate change and higher water O. Vileikis and L. Fakhoury, Risk Manage-
levels. While with the ABC method one constantly has to aggregate and dis- ment at Heritage Sites: A Case Study of the
aggregate to find the comfortable level at which to quantify the risk, Petra World Heritage Site (Paris/Amman:
CPRAM allows for adding the magnitudes of risk of specific scenarios to UNESCO, 2012), http://unesdoc.unesco.
org/images/0021/002171/217107 m.pdf
express the magnitude of a generic risk or number of similar scenarios.
(accessed 3 September 2015); Waller, ‘Con-
They are deterministic in the sense that each scenario has a single servation Risk Assessment’, 12–16; and
outcome, although best, worst and most likely case options may be given ‘Special Issue: International Symposium
to indicate the range of the uncertainty. on Cultural Property Risk Analysis’, ed.
With these methods probability and consequence can be described in Robert Waller, Collections: A Journal for
Museum and Archives Professionals 8, no. 4
terms of low, medium or high, without being too exact about what each cat-
(2012) and 9, no. 1 (2013).
egory means. Results can then be represented in a risk matrix (Fig. 1 left) to
illustrate and communicate the outcomes of any assessment. Instead of 4 R. Waller and S. Michalski, ‘A Para-
digm Shift for Preventive Conservation,
named categories, probability and consequence can be assessed on a and a Software Tool to Facilitate the
scale of 1 – 5 which allows for calculating a composite risk index which Transition’, in Preprints of the ICOM–
then allows for ranking risks in a bar graph (Fig. 1 middle). Alternatively, CC 14th Triennial Meeting, 733–8.
probability and consequence may be expressed in numbers to calculate 5 J. Ashley-Smith, Risk Assessment for
the expected loss (for example, in monetary units or percentage of the Object Conservation (Oxford: Butter-
asset) within a certain period of time (Fig. 1 right). The ABC method uses worth-Heinemann, 1999).
logarithmic scores, whereas CPRAM calculates the fraction of the collection 6 See, American Society of Heating,
lost in a century. Graphs serve as good illustrations, show rankings and are Refrigerating and Air-Conditioning
generally well understood by colleagues at all levels and so allow for dis- Engineers, ‘Museums, Libraries and
cussion of the outcome. Archives’, in ASHRAE Handbook: Heating,

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

20 Brokerhof and Bülow

Ventilating, and Air-Conditioning Appli-

cations, SI Edition (Atlanta: ASHRAE,
2003), 21.1–21.23; B. Ankersmit, Klimaat-
werk—Richtlijnen voor het museale binnenkli-
maat (Amsterdam: Instituut Collectie
Nederland and Amsterdam University
Press, 2009); Commission Internationale
de l’Eclairage, Control of Damage to
Museum Objects by Optical Radiation, Tech-
nical Report CIE 157:2004 (Vienna: Com-
Fig. 1 Typical examples of graphical representations of outcomes of risk assessments: risk
mission Internationale de l’Eclairage,
matrix (left), bar graph as seen with the ABC method (middle) and 3D bar graph as seen
2004); A.W. Brokerhof, Het beperken van
with CPRAM (right).
lichtschade aan museale objecten: Lichtlijnen
(Amsterdam: Instituut Collectie Neder-
land, 2005), http://cultureelerfgoed.nl/
sites/default/files/publications/ Obviously the challenge in a scenario-based approach is to be compre-
informatieblad_13_beperken_ hensive and think of all the relevant scenarios and then analyze them.
lichtschade_museale_objecten.pdf The effort demanded forms a major hurdle for application.
(accessed 2 September 2015); and British Another method to visualize risk is a risk map. Risk maps originate from
Standards Institution, Specification for
and are used in geological and geographical areas of risk management,
Managing Environmental Conditions for Cul-
tural Collections, Publicly Available Specifi- such as for earthquake and flood risks. Like weather maps, colour codes
cation PAS 198:2012 (London: British indicate probabilities, impacts or both. Without describing specific scen-
Standards Institution, 2012). arios, it is easy to see where exposure is expected or losses may occur.
7 See for example, Elkin et al., ‘A Database This method is not uncommon to show risks in archaeology and built heri-
Tool for Collections Risk Evaluation and tage.8
Planning’; J. Subic Prislan, E. Cerar and The risk mapping approach has been adapted for interiors and collec-
V. Zivkovic, ‘Who Cares? We Do—A tions.9 For example, to get a quick insight into risks from light in a room,
Nationwide Survey of Museum Storage
maps showing the distribution of the collection throughout the room are
in Slovenia’, in Bridgland, ICOM-CC
17th Triennial Conference Preprints, overlaid with maps of the value distribution, vulnerability to light and
https://www.academia.edu/11635569/ exposure to light. The resulting compilation shows where vulnerable
Who_cares_We_do_A_nationwide_ objects with high value are exposed to high light levels, and hence where
survey_of_museum_storage_in_Slovenia the biggest losses of value can be expected (Fig. 2). A more detailed analysis
(accessed 3 September 2015); and A.W.
for those objects may confirm this and appropriate measures can be taken,
Brokerhof, B. Ankersmit and
F. Ligterink, Digitale Handboek Collectie- such as reducing light levels or moving objects, without a complete risk
Risicomanagement (Amersfoort: Cultural analysis of the entire collection in the room.
Heritage Agency of the Netherlands,
2013), http://cultureelerfgoed.nl/
dossiers/collectierisicomanagement
(accessed 3 September 2015).
8 See for example, G. Accardo, E. Giani
and A. Giovagnoli, ‘The Risk Map of
Italian Cultural Heritage’, Journal of Archi-
tectural Conservation 9, no. 2 (2003): 41–57;
European–Mediterranean Seismological
Centre, ‘History’, European–Mediterra-
nean Seismological Centre, https://
www.emsc-csem.org/about/ (accessed
24 January 2016); Risicokaart.nl, ‘Risico-
kaart’, Interprovinciaal Overleg, Minist-
erie van Binnenlandse Zaken en
Koninkrijksrelaties, Ministerie van Infra-
structuur en Milieu, 2015, http://www.
risicokaart.nl/en/ (accessed 24 January
2016); FEMA, Understanding Your Risks:
Identifying Hazards and Estimating Losses,
State and Local Mitigation Planning
How-to Guide 386-2 (Washington, DC:
FEMA, 2001), http://www.fema.gov/
media-library/assets/documents/4241
(accessed 28 July 2015); and S. Tyagunov,
G. Grünthal, R. Wahlström,
L. Stempniewski and J. Zschau, ‘Seismic Fig. 2 Example of a ‘risk map’: superimposing maps of an interior with colour coding for cul-
Risk Mapping for Germany’, Natural tural value, vulnerability and exposure to light combined with the distribution of the risk of
fading throughout the room.

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

The QuiskScan—a quick risk scan to identify value and hazards in a collection 21

Mapping a collection—flying over instead of walking through Hazards and Earth System Sciences 6 (2006):
A scenario-based approach would not provide the required outcome within 573–86, https://hal.archives-ouvertes.
the given resources for the museums that the authors have worked with. fr/hal-00299344/document (accessed 28
Instead, a variation of risk mapping of the entire collection seemed, in July 2015).
their experience, to be the direction in which to proceed. Thus the risk 9 A.W. Brokerhof, ‘Exhibiting with
mapping approach was developed into a matrix-based scan; a ‘mapping’ Minimal Light Risks’ (presentation at
of the collection within a matrix. Similar to overlaying geological maps of ‘The Future’s Bright: Managing Colour
Change in Light Sensitive Collections’,
asset values, vulnerability and exposure, the matrix crosses value (signifi- National Museum, Stockholm, Sweden,
cance) of collection units with vulnerability to the 10 agents of deterio- 15 November 2012).
ration, thus identifying a ‘vulnerable value’. If desired, this could be
overlapped with, or analyzed for exposure, to get an indication of the mag-
nitude of risk. This approach allows for a quick risk scan for the collection,
hence its name QuiskScan.
One might think of this methodology as flying over a forest instead of
walking through it.10 If each tree were to represent a risk, its height 10 A.W. Brokerhof, ‘Risk Assessment
would be a measure of its magnitude. Without a perspective to see the for Cultural Heritage—Methods and
Tools’ (presentation at the Konferens
height from ground level, one needs to measure all the trees to determine
Riskhantering och kulturvård, Saltsjö-
which ones are the tallest. This implies measuring more trees than necess- baden, Stockholm, Swedish Heritage
ary to guide further decisions. A benefit of this approach is that one can Board, 1–3 December 2014), https://
argue that the organization is managing the risks in areas where trees are youtu.be/2zbk_nFzFl0 (accessed 2 Sep-
short. Nevertheless, the forester with limited time wants to have an idea tember 2015).
of where the tall trees are and zoom in on those to see whether they
require action without spending time on measuring all of the trees. When
risk is considered to be the probability of loss, then the biggest losses
will occur where high value meets high vulnerability. Hence for a quick
(preliminary) risk assessment, one wants to find the ‘vulnerable value’
(Fig. 3). Vulnerability in this case focusses on the tangible susceptibility of
an object or collection (unit) to undergoing material change as a result of
exposure to an agent of deterioration. In addition, susceptibility to intangi-
ble factors such as political, social, economic, and reputational, can be con-
sidered when appropriate.11 11 O.D. Cardona, ‘The Need for
Rethinking the Concepts of Vulner-
ability and Risk from a Holistic Perspec-
tive: A Necessary Review and Criticism
for Effective Risk Management’, in
Mapping Vulnerability: Disasters, Develop-
ment and People, ed. G. Bankoff, G. Frerks
and D. Hilhorst (London: Earthscan
Publishers, 2004), 37– 51.

Fig. 3 Difference between a scenario-based approach, walking through the forest (upper) and
the QuiskScan, flying over the forest (lower).

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

22 Brokerhof and Bülow

In the following section we illustrate the QuiskScan approach as applied

to a fictional collection, discuss its advantages and disadvantages, and then
reflect on using the approach in practice.

The QuiskScan approach

The QuiskScan combines two tools that have been developed in recent years
by the Cultural Heritage Agency of The Netherlands (RCE): Assessing
12 T. Luger, A.W. Brokerhof, S. Hartog Museum Collections and Collection Risk Management.12 Both tools are
and G. Huisman, Assessing Museum Col- applied in a simple form by assessing value and vulnerability of collection
lections; Collection Valuation in Six Steps
units relative to each other, ranking them amongst each other rather than
(Amersfoort: Cultural Heritage Agency
of the Netherlands, 2014), http:// aiming for absolute assessments and comprehensive description. In line
cultureelerfgoed.nl/publicaties/ with both of these tools, the QuiskScan proceeds through a number of
assessing-museum-collections (accessed steps: determining the context, drawing up the collection anatomy, asses-
2 September 2015); Brokerhof, Ankersmit sing the relative value of the collection units, determining their vulner-
and Ligterink, Digitale Handboek Collectie- ability for the agents of deterioration, and considering the levels of
Risicomanagement.
exposure. The end result is a matrix as sketched in Table 1. Each step is dis-
cussed in further detail below.

1 Setting the context: purpose, scope, and collection anatomy

Value assessment and risk assessment are preferably processes that are con-
ducted with a team, with the context in which the process takes place
understood and agreed by all team members.

1a Purpose
The purpose should be clear to all involved. Why is the collection assessed?
What questions need to be answered or which decisions need to be made
that require insights into collection, values, and risks? Having a clear
picture of the purpose also allows limiting the scan to the relevant collection
units and agents of deterioration and thereby avoids unnecessary work.

1b Scope
The scope should be defined. Which part of the collection is being scanned?
This is closely related to the purpose. If decisions do not involve the entire
collection, then the scope can be limited to the relevant parts. For example,
within a paper-based collection, one might only want to examine the prints
and drawings, and exclude libraries and archives. The scan can always be
expanded at a later stage if required. The scope is also needed to determine
the reference framework for value assessment. Is the scope the entire collec-
tion or a few units within that collection? Are values to be assessed within a
regional, national or even international context? This is needed when it
comes to determining what makes a collection score ‘high’ or ‘low’ for
the various criteria. It provides a level of comparison.

Table 1 Result of the QuiskScan showing the collection anatomy, the value of collection units
A– E with a weighting and the vulnerability of each unit for the agents of deterioration I– V.

Notes: H¼high, M¼medium, L¼low. Colour codes: red¼H×H, orange¼H×M¼M×H, yellow¼M×M

green¼M×L¼L×M¼H×L¼L×H, blue¼L×L.

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

The QuiskScan—a quick risk scan to identify value and hazards in a collection 23

1c Collection anatomy
Considering the entire collection as one unit for the scan would not give a
useful output for decision making. Thus it is necessary to split the collection
into units. This subdivision is referred to as the collection anatomy. The
more units, the more work the group will need to do. Hence when deter-
mining the collection anatomy one aims at having as few units as possible,
yet at a level of differentiation such that the objects within each collection
unit show a high degree of similarity and are well distinguished from
other units. Collection units can be based on, for example, curatorial classi-
fication, object type, material or location. For each collection unit, its facts
and figures, a short description and its quantity in some logical dimension
are listed. For example, the number of objects or boxes, meters of shelving,
volume, weight or bytes.
Like any value or risk assessment, the process should be well documen-
ted so that its arguments and decisions can be traced back and understood
by custodians in the future.

2 Assessing current value

In order to assess value the following questions are typically asked:

Why are the collection units kept?

What is their significance?

Which values do we attribute to them?

How high are those values?

The collection’s units are compared with each other within the context of
the host organization’s constitution or mission. Units are given an overall
‘high’, ‘medium’ or ‘low’ score, based on arguments from their evaluation
against a standard set of criteria described in the publication, Assessing
Museum Collections; Collection Valuation in Six Steps.13 13 Luger et al., Assessing Museum Collec-
The method distinguishes between attributes and value criteria. Attri- tions.
butes consider state and condition, completeness, provenance and rarity.
They do not in themselves make something valuable, but can enhance
value when they are good or attenuate the value when they are poor.
Units can score ‘good’, ‘average’ or ‘poor’ for these attributes.
The value criteria are divided into three clusters: cultural-historical (artis-
tic, historic, information values), social and societal (contemporary societal
and personal experience values), and use (values related to the museum’s
business and its public). Objects or collection units can score ‘high’,
‘medium’ or ‘low’ for these criteria.
In order to assess the units consistently, the team has to agree on the
evaluation framework and define for each criteria what makes a unit
score ‘high’ or ‘low’, ‘good’ or ‘poor’. Once the framework is available,
the assessment can be made.
The assessment of attributes and value criteria combines to create a
‘value score’ for each collection unit which is entered under ‘current
value’ in the matrix for each collection unit, as in the example of Table 1.
This is not a calculated sum of individual scores. It is the overall value jud-
gement underpinned by the arguments set out in the agreed evaluation fra-
mework.
In practice it proves to be easier to start with the team’s overall score and
ask them to provide arguments for that assessment by going through the
criteria rather than by constructing a final verdict based on the accumu-

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

24 Brokerhof and Bülow

Fig. 4 Distribution of total collection value from Table 1 over value categories visualized in a pie
chart (value pie) (a, upper left) and a weighted value pie (b, upper right), in a mosaic plot (c,
lower left) and a population distribution (d, lower right).

lation of all the details. For example, curators are usually aware of what has
a higher or lower value, yet are rarely asked to make such evaluations expli-
cit. With the QuiskScan the criteria can be used to support arguments so as
to help reach overall agreement amongst the diverse range of specialists in
the team. For example, within the assessed prints and drawings collection
one might agree that the watercolours by J.M.W. Turner are of high value
based on their artistic merits, their historic significance and their popularity
with the public. The final verdict is usually reached through a mix of
zooming in and zooming out between general impressions and detailed
arguments. Furthermore, if a collection unit contains objects with different
overall values, then, in our opinion, the most pragmatic way to deal with
this is to attribute the mode or median value as an overall judgement
whilst specifying any exceptions for later consideration. This is where the
team’s moderator plays an important role in soliciting arguments and rep-
resentations in order to arrive at a group decision, with the caveat that this
can be subsequently amended, if deemed appropriate.
Based on these overall scores the collection anatomy thus provides
insight into how the total collection value is distributed between the
various collection units. There are several different ways of visualizing
this. Figure 4 shows examples of a value pie (a), weighted value pie (b),
mosaic plot (c) and population distribution (d) for the example of Table 1.
While in the value pie (a) the number of objects in each value category is
plotted, the distribution of the total collection value in the weighted
value pie (b) shows where most of the value actually is. In this case the
‘medium’ value is normalized at 1, with ‘high’ value objects being 10
times as valuable while ‘low’ value objects are just 0.1 times as valuable.
In terms of equivalent value, one would swap for instance one ‘high’
value object for 10 objects of ‘medium’ value and for 100 ‘low’ value
objects. In the example of Table 1 the 1800 objects of units D and E only con-

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

The QuiskScan—a quick risk scan to identify value and hazards in a collection 25

tribute 1% to the total collection value, whereas the 1500 objects in units A
and B together account for 74% of the total collection value.

3 QuiskScan—finding vulnerable value and risks

The quick risk scan provides insight into the hazards to the collection and
identifies possibilities for loss of value. This is done in two steps. The first
step is to scan the vulnerability of the collection units to each of the agents
of deterioration. In line with the value assessment, collection units may
contain objects with different vulnerabilities, and again the most pragmatic
way to deal with this is to give the mode or median vulnerability and then
specify any exceptions for later consideration. As with all assessments, one
member of the team documents the discussions and conclusions made
about vulnerability while the moderator solicits all the relevant arguments.
The verdicts reached on the level of vulnerability each unit has to each
agent of deterioration are entered in the matrix for the collection units as
‘high’, ‘medium’ (average) or ‘low’, and colour codes can be used to classify
the overall ‘vulnerable value’. In the example in Table 1 this is shown
for five unspecified agents of deterioration. The combination (High
value)×(High vulnerability) would result in the biggest loss of value to
the collection if found to be exposed to the agent and is coded red
(HH¼red). For example, a collection of Turner watercolours would be
highly vulnerable to water and would be coded red. Collection units
having (High value)×(Medium vulnerability) or (Medium value)×(High
vulnerability) would result in a slightly smaller loss of value if
exposed and are coded orange (HM¼MH¼orange), while (Medium
value)×(Medium vulnerability) becomes yellow (MM¼yellow). As soon
as either the value or the vulnerability is low, the potential loss of value
to the collection becomes small and its cell in the matrix is coded green
(HL¼LH¼ML¼LM¼green). The combination of (Low value)×(Low vul-
nerability) will give a very small loss of value and is coded blue (LL¼blue).
Having assessed ‘vulnerable value’, the next step is to consider whether
there is exposure at all. For example, if there are no sources of water near
the Turner watercolours, there is no water risk to them either. However,
where exposure is expected further risk assessment requires a scenario-
based approach. The QuiskScan has served its purpose. It has highlighted
the relevant areas for further analysis. For the Turner watercolour
example, this means identifying the cause – pathway – effect scenarios for
water, analysing probability and consequence, and evaluating their magni-
tude of risk.

Results of the scan and their interpretation

The results of the QuiskScan can provide various insights into the risks
facing a collection. In the example of Table 1, collection units A and B
stand out as areas where significant loss of value to the collection may
occur (denoted by the red cells in the matrix) because of their high value
and high vulnerability to some of the five unspecified agents of deterio-
ration. While both are vulnerable to agent I, unit B is also vulnerable to
agents IV and V, and for these four combinations one would want to
know more about the levels of exposure to ascertain the probability of
loss of value. At this point the team needs to consider scenarios that may
lead to exposure. If there is no exposure, for example because there are
already mitigation measures in place, the actual risk is smaller and may
not require further attention. Since both units A and B are vulnerable to
agent I, it may be effective to focus on that agent and reduce the exposure
as this reduces the risk to the majority of the collection value in one action.
Since collection unit B faces risks from three agents, it may be useful to
analyse what can be done to reduce exposure to all three in one action

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

26 Brokerhof and Bülow

Fig. 5 Vulnerability of the collection from Table 1 for agents of deterioration I-V stacked as
number of objects per collection unit classified as ‘high’ (red), ‘medium’ (yellow) and ‘low’
(green) (left) and multiplied by their value to classify the ‘vulnerable value’ in terms of a per-
centage of the collection for each agent (right).

by, for example, moving the unit somewhere else, or improving specific
conditions.
From the number of objects in each collection unit and the mode or
median assigned for their value and vulnerability to the different agents
one can generate broad brush profiles for the entire collection. Figure 5
shows the vulnerability of the collection to each agent of deterioration as
number of objects classified as ‘high’, ‘medium’ and ‘low’ vulnerability
(left). It can be seen that some 80% of the collection has a high vulnerability
to agent I. If that 80% is also exposed to that agent, this provides convincing
arguments to at least pay attention to the agent and its mitigation. To a
slightly lesser extent this also holds for agents III and IV. Agent V seems
less of a threat as only unit B with its 1000 objects shows a high vulner-
ability. This is where the inclusion of the value becomes important.
Figure 5 (right) shows the stacked graph for percentage ‘vulnerable
value’ of the collection. Now agent V appears as relatively more important
than III which necessitates further analysis. The matrix and the graphs
become the basis for discussion about further action, such as the need for
a more detailed analysis of exposure, the probability of damage and the
estimation of actual loss of value. After all, not all damage results in a sig-
nificant loss of value. It is also at this point that the exceptions, flagged up
earlier in the process, need to be reconsidered to assess their impact on the
overall picture and whether they deserve special attention.
At this point it can be seen that the QuiskScan can be used to form a sys-
tematic overview of the collection and its risks, but that it contains a con-
siderable degree of uncertainty due to the high level of aggregation and
the possible variation within any particular set of collection units. Thus,
whilst the QuiskScan methodology can be said to have some limitations,
it can form the basis for prompting a more in-depth risk analysis using
one of the established methods in the areas of the collection where it has
indicated the possibility of significant loss of value.

Discussion
The QuiskScan should be regarded as a tool that fits in between relying on
best practices and conducting a comprehensive risk assessment. Due to its
broad brush approach many details will be missed, and as with other
models and tools, the output is only as good as the input. Weaknesses in
the QuiskScan include its initial high level of aggregation, a disregard for
the variability in both value and vulnerability within collection units by
basing its assessment methodology on modes and medians, and its accep-
tance of large uncertainties. Furthermore, as the QuiskScan attempts to sys-
tematically extract data from existing knowledge in the first instance, there

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

The QuiskScan—a quick risk scan to identify value and hazards in a collection 27

is a tendency towards a biased input and therefore a biased output. Low

probability risks may be overseen, impacts or probabilities underestimated,
vulnerability of collections may be unknown and operators’ levels of confi-
dence and control may be overestimated, while only obvious and known
risks are considered.
Yet, arguably, all these disadvantages are outweighed by a number of
advantages. The QuiskScan can generate an integral overview of the collec-
tion, its values, vulnerability and risks within one or two group sessions.
Moreover, the bias can be reduced by an experienced moderator. The fact
that the process does not require generating hitherto unknown data in
the first instance means that team members are more easily engaged with
the process. They contribute with knowledge, experience and data avail-
able to them and participate in discussions that produce group verdicts
in a relatively short time. This in turn results in an early buy-in, enthusiasm
to continue and a sense of achievement in using the methodology. The
assessment is an iterative process, consolidating what the group knows
and revealing what is unknown. The process allows for revision and refine-
ment, and thus forms a solid basis for discussions and shared insight into
the collection.
At the time of writing curatorial departments at the British Museum have
started to apply the approach to produce a shared overview of their collec-
tions, and their values and vulnerabilities have been discussed and
recorded. In the case of the Department of Prints and Drawings this has
led to the setting of shared priorities with the Department of Conservation
and to the development of a new work programme.14 At RCE the QuiskScan 14 A.E. Bülow, A.W. Brokerhof, C. Barry
is used as a tool in consultancy and for group sessions with museum staff and H. Chapman ‘Pride and Prejudice:
Developing a Shared Understanding of
in, for example, assessing current situations and needs for improvement in
Priorities’, in Abstracts of the XIII IADA
collection care and to help draw up specifications for new exhibitions and Congress, October 2015, Berlin (Berlin:
new buildings. IADA, 2015), 70.
Finally, the QuiskScan can be integrated into the Value Management Scan,
an approach developed by RCE in cooperation with the Amsterdam City
Archives, that looks not only at possible losses, but also at possible gains
of value.15 15 J. Kemp, A. Bülow and A. Brokerhof,
‘Value Management Scan: Setting Priori-
ties in Management and Care of Collec-
tions’, in Abstracts of the XIII IADA
Conclusion Congress, October 2015, Berlin (Berlin:
While the development of the QuiskScan continues, the authors have found IADA, 2 015), 95.
its application to meet the challenge of generating an overview of a collec-
tion and its risks within a short period of time. Experience has shown that it
quickly creates shared risk awareness and serves as a lever to more in-
depth risk analysis. It reduces the apprehension to embark on a risk assess-
ment project and builds a bridge between letting best practice guide
decisions and comprehensive risk analysis for which it provides the basis.

Acknowledgements ation’s approach to the care of its collection, there remains a

The authors would like to thank all colleagues at the organizations general apprehension concerning the significant effort required to
where the QuiskScan was applied for their contributions to its devel- execute such an assessment. This paper describes the development
opment. Particular recognition goes to Dr Hugo Chapman, Keeper of the QuiskScan, a quick risk scan, which yields an overview over
of Prints and Drawings, British Museum, London and Dr Bart a collection, its values and vulnerabilities with comparatively little
Ankersmit, senior scientist, Conservation and Restoration Depart- effort. The QuiskScan uses a matrix-based approach to map value
ment, Cultural Heritage Agency of The Netherlands, Amsterdam and vulnerability to the agents of deterioration for different collec-
for their input and support and to the reviewers of this paper for tion units to highlight where significant losses to the collection
their constructive feedback. might occur. Like other risk assessment approaches the QuiskScan
involves expert input from across the organization thus helping to
Abstract create a shared insight into the collection, and an institutional
Over the past two decades a number of heritage institutions have awareness of risks. The method should be regarded as a tool that
conducted different forms of risk assessments. Despite the generally fits in between relying on best practice and conducting a compre-
agreed profound changes such assessments make to an organiz- hensive risk assessment.

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016

28 Brokerhof and Bülow

Résumé evaluaciones consiguen cambiar la manera en la que dichas organi-

«Le QuiskScan—une analyse des risques rapide pour identifier la zaciones se plantean el cuidado de sus colecciones, sigue habiendo
valeur et les dangers au sein d’une collection» una aprehensión general respecto al esfuerzo significativo que
Au cours des deux dernières décennies, un certain nombre d’insti- supone el llevar a cabo este tipo de evaluación. En este trabajo se
tutions patrimoniales ont mené différentes formes d’évaluations describe el desarrollo del QuiskScan, un rápido análisis de riesgos
des risques. Bien qu’il soit généralement reconnu que de telles que da una visión general sobre una colección, sus valores y vulner-
évaluations amènent de profonds changements dans la manière abilidades, con relativamente poco esfuerzo. El QuiskScan utiliza un
dont une organisation considère la conservation de sa collection, il planteamiento basado en matrices; asignando el valor y la vulner-
reste une appréhension générale face à l’effort considérable requis abilidad con respecto a los distintos agentes de deterioro en los dis-
pour réaliser une telle évaluation. Cet article décrit le développe- tintos sectores de la colección se resalta donde podrı́an producirse
ment de la QuiskScan, une analyse rapide des risques, qui donne pérdidas significativas de la colección. Al igual que otros análisis
un aperçu d’ensemble sur une collection, ses valeurs et ses vulnér- de riesgo, y con el fin de crear una visión común de la colección y
abilités avec relativement peu d’effort. Le QuiskScan utilise une una conciencia institucional de riesgos, el QuiskScan requiere la par-
approche fondée sur une matrice permettant de cartographier la ticipación de expertos de toda la organización. Este método debe ser
valeur et la vulnérabilité de différents objets de la collection face considerado como un instrumento que ocupa un lugar intermedio
aux agents de détérioration et de mettre en évidence les pertes entre basarse exclusivamente en las mejores prácticas y llevar a
importantes que la collection peut subir. Comme d’autres méthodes cabo una evaluación de riesgos exhaustiva.
d’évaluations des risques, l’approche de la QuiskScan implique
l’implication d’experts de toute l’organisation, aidant ainsi à créer
un aperçu commun de la collection et une prise de conscience insti-
tutionnelle des risques. La méthode doit être considérée comme un
outil intermédiaire entre les meilleures pratiques et une évaluation
complète des risques.

Zusammenfassung
„Der QuiskScan-ein schnellerScan um Werte und Risiken in einer
Sammlung zu identifizieren“
Während der letzten zwei Jahrzehnte haben eine Reihe von Kultur- Biographies
institutionen verschiedene Formen von Risikoanalyse durchgeführt. Agnes Brokerhof studied Chemistry and Art History at Leiden Uni-
Obwohl der Konsens ist, dass solche Analysen tiefgreifende Verän- versity (The Netherlands). From 1989 to 1991 she worked at the
derungen bei der Sammlungspflege innerhalb einer Institution Australian Museum in Sydney and was a research fellow at
auslösen, besteht ein allgemeiner Widerstand gegen den nicht uner- CSIRO in Canberra (Australia). After taking the ICCROM course
heblichen Aufwand, den die Durchführung einer solchen Analyse ‘Scientific Principles of Conservation’ in Rome (1992) she joined
mit sich bringt. Dieser Artikel beschreibt die Entwicklung des the Central Research Laboratory for Objects of Art and Science in
‘QuiskScan’, einer schnellen Risikoanalyse, die mit vergleichsweise Amsterdam as conservation scientist. Subsequently she worked at
wenig Aufwand einen Überblick über eine Sammlung, ihre Werte, ‘Instituut Collectie Nederland’ (ICN) as programme manager of
und ihre Schadensanfälligkeit gibt. Der QuiskScan nutzt eine ‘Collection Risk Management’. Since the merger of ICN into the
Methode, die auf einer Matrix basiert, mittels derer Werte und Cultural Heritage Agency of the Netherlands (RCE) in 2011 she
Schadensanfälligkeit unterschiedlicher Sammlungsteile gegenüber has been a senior conservation scientist in its Conservation & Res-
verschiedenen Risiken identifiziert werden können. Ähnlich toration Department based in Amsterdam. Currently her activities
anderer Analysemethoden bedingt der QuiskScan den Einsatz rele- focus on teaching, coaching and applying the developed models for
vanter Experten innerhalb einer Institution, was bei der Bildung value and risk assessment in collection management.
gemeinsamer Ziele und Wissen um die Risiken einer Sammlung Dr Anna Bülow apprenticed as a bookbinder in Germany before
hilft. Die Methode sollte als ein Werkzeug verstanden werden, das graduating with a degree in paper conservation from the
zwischen bewährten Praktiken und einer umfassenden Risikoana- Hochschule für Gestaltung, Kunst und Konservierung in Bern, Swit-
lyse liegt. zerland. She worked as a research fellow at the Canadian Conserva-
tion Institute in Ottawa and graduated from Queen’s University in
Resumen Kingston, Ontario, Canada, with an MAC in conservation research
“El QuiskScan—un rápido análisis de riesgos para identificar el in 1999. Specializing in preventive conservation, she completed
valor y las amenazas de una colección” her PhD at De Montfort University in Leicester, UK, in 2002, and
En las últimas dos décadas, diferentes instituciones que se ocupan worked as Head of Preservation at The National Archives, UK
del patrimonio han llevado a cabo distintos tipos de evaluaciones between 2003 and 2013. She is currently the Head of Conservation
de riesgos. A pesar de que en general hay consenso de cómo tales at the British Museum.

Contact address
Agnes W. Brokerhof Dr Anna E. Bülow
Senior Conservation Scientist Head of Conservation
Conservation & Restoration Department British Museum
Cultural Heritage Agency of The Nether- Great Russell Street
lands London WC1B 3DG
Hobbemastraat 22 UK
1071 ZC Amsterdam, The Netherlands Email: [email protected]
Email: [email protected]

Journal of the Institute of Conservation, Vol. 39 No. 1 March 2016