Situation-aware Building Information Models for Next Generation
Building Management Systems
Ovidiu Noran
1,2 a
, Peter Bernus
1b
and Sorin Caluianu
2c
1
IIIS Centre for Enterprise Architecture Research and Management - Griffith University, Brisbane, Australia
2
Faculty of Installations Engineering, Technical University of Constructions Bucharest, Romania
Keywords: Building Information Model, Building Management System of Systems, Decision Support Systems,
Situational Awareness, Big Data, Data Warehousing, Decision Model, Situated Reasoning, Channel Theory.
Abstract: Technical advances in Information and Communication Technology have enabled the collection and storage
of large amounts of data, rising hopes of improving asset decision-making and related building management
support systems. It appears however that the gap between the required decision-making knowledge and the
actual useful information provided by current technologies appears to increase, rather than contract. Thus,
often the multitude of patterns afforded by current data analytics techniques does not deliver a set of scenarios
prone to effective decision making. This paper advocates a decision analytics solution featuring the use of
Situated Logic to create ‘narratives’ providing adequate meaning to data analytics results, and the use of
Channel Theory so as to support adequate situational awareness. This approach is also analysed in the context
of a Building Management System-of-Systems paradigm, highly relevant to the emerging complex Clusters
of Intelligent Buildings within Smart Cities, featuring collaborative decision-making centres and their
associated decision support systems.
1 INTRODUCTION
Building information models (BIMs) are files
typically containing proprietary formats and data
which can be extracted, exchanged or networked to
support decision-making within a Building
Management System (BMS) for a built asset (Van
Nederveen & Tolman, 1992).
BIM can be used to enhance building decision
making (Nowak, Ksiazek, Draps, & Zawistowski,
2016), to coordinate projects (Rokoeei, 2015) and
deliver sustainable building value (Fadeyi, 2017).
BIMs have an important role in reducing the
fragmentation among professionals at each stage and
across building delivery stages by providing a virtual
repository that allows easy access to- and sharing of
information and knowledge in real time. Thus, BIM
also enhances interoperability, providing a platform
for professionals to work in an integrated
environment at any stage of the building delivery
a
https://orcid.org/0000-0002-2135-8533
b
https://orcid.org/0000-0001-5371-8743
c
https://orcid.org/0000-0001-5961-4187
process and further on in the management phase
within BMS (Santos, 2009).
Although traditionally focused more on the short term
and real-time operation, BIMs and BMS are
increasingly required to assist in medium- and long
term asset planning, especially in the case of
Intelligent Building Clusters in Smart Cities (Zhang
et al., 2018). The scope of the models is also expected
to expand from mainly energy consumption to all
aspects of asset management so as to provide a
holistic and integrated view (Teliceanu, Golovanov,
Lazaroiu, & Dumbrava, 2017; Wei & Zhu, 2016). It
must also be noted that nowadays, due to the
increased complexity brought by numbers and variety
of controlled parameters, BMS are increasingly
taking on hybrid agent features, featuring human and
machine components, beyond mere Human Machine
Interface (HMI) within the BMS ‘head end’ (Forte
Asset Services, 2019). All of the above endeavours
essentially depend on the quality and relevant content
of the information obtained from the collected data.
Noran, O., Bernus, P. and Caluianu, S.
Situation-aware Building Information Models for Next Generation Building Management Systems.
DOI: 10.5220/0008767900750082
In Proceedings of the 22nd International Conference on Enterprise Information Systems (ICEIS 2020) - Volume 1, pages 75-82
ISBN: 978-989-758-423-7
Copyright
c
2020 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
75
Thus, Big Data (Li, Zhang, Hu, & 17., 2017), the
Liquid Enterprise (Bayler, 2016), Sensing
Information Systems (Zdravković, Noran, &
Trajanović, 2014) and similar concepts hold the
promise to provide all necessary decision-making
information for building management in the adequate
detail, quality and ‘freshness’ required. Conceptually,
this endeavour comprises achieving the necessary
capabilities to use the data to derive decision-making
information in an efficient and effective manner,
based on inferring knowledge that was not available
(and attainable) before; nowadays, this assumes the
presence and support of suitable data analytics.
The degree of success in supporting decision-
making heavily depends on proper data synthesis and
interpretation (Dibb, Meadows, & Wilson, 2015).
Thus, proper data analytics will effectively support
decision analytics.
It is also becoming clear that finding new ways to
correctly interpret complex data in context is
necessary (SAS, 2015). For example, evidence-based
medicine that relies on large scale data gathering
through clinical trials and careful statistical analysis
is now showing difficulties when the evidence
gathered is applied in complex individual cases
(Greenhalgh, Howick, & Maskrey, 2014; Pope,
2003).
From the above, an obvious perspective is that
when intending to use large amounts of gathered data
to create useful decision-making information, one
must carefully consider the information needs of the
intended audience (e.g. management) and
importantly, how the interpretation of data is
influenced by context.
This paper intends to analyse the barriers in using
data warehousing and big data approaches for BIM in
order to support proper decision analytics manifested
in effective decision-making within BMS.
2 DATA TO INFORMATION:
CONCEPTS AND
APPROACHES
2.1 Building Management Systems and
Their Information Needs
Initially limited to operational and real time, Building
Management Systems nowadays increasingly need to
also make decisions on strategic and tactical levels.
This endeavour can be reasoned about in relation to
the information flow, usage and needs, and optimised
using various types of models. In the following, the
authors will use an example of mainstream systematic
model of decision-making, namely the GRAI Grid
(Doumeingts, Vallespir, & Chen, 2003) (see Fig. 1).
Fundamentally, this generic model identifies
management, command and control tasks at various
levels (identified via time spans called ‘horizons’)
and the information flow between them.
The exogenous and endogenous information
flows feeding the Manage, Command and Control
centre in Fig. 1 illustrate the point that in order to
make successful decisions it is necessary to satisfy the
information needs of the management functions.
Figure 1: Information flows in a Basic mainstream decision
making model: GRAI Grid (Doumeingts, Vallespir, &
Chen, 2003).
This means that the data gathered and analysed
must be meaningful, properly aggregated (level of
detail) and suitably expressed in order to meet the
demands and competencies of each audience
populating the decision centres at various horizons.
This is not a trivial task. To justify this point the
authors refer to two main approaches to data
analytics, namely data warehousing and big data.
2.2 Data Warehousing and Big Data
The concept of Data warehousing is relying on high
quality clean and integrated data to enable the use of
operational databases and other data repositories
snapshots and build an interface enabling the analysis
(‘mining’) in order to identify management-relevant
information. Some authors (Inmon, Zachman, &
Geiger, 1997; Kimball, 1996) argue that such a
facility could be built fast and in an affordable manner
using existing databases and possibly transaction logs
so as to gain management insight (Matte & Rizzi,
2009; Satyanarayana, Srinivasu, Poorna Rao, &
Rikkula, 2010). The aim is to create a narrative that
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
76
is characterising the present or predicted future
situation and is essential for decision-making. After
some initial success in creating meaningful insight for
management, data warehousing displayed some
notable failures to deliver on its promises (Adelman
& Moss, 2000).
‘Big data’ is a technology that uses traditional data
analysis and machine learning techniques to derive
useful interpretations based on large and varied data
sources (Gandomi & Haider, 2015; Marr, 2015).
Brought forward by the technological advances in
data gathering (cheaper and more intelligent Internet-
of-Things (IoT)-enabled sensors, cheaper and cloud-
based storage etc.) and initially boasted as the
solution to the problems where traditionally
implemented data warehousing fell short, big data is
still maturing and yet to make significant inroads in
decision support (Horitaa, de Albuquerquea,
Marchezinid, & Mendiondoc, 2016; Kościelniaka &
Putoa, 2015). This is also due to its inherent
dependence on machine learning algorithms that
attempt to predict but cannot adequately explain the
predictions, which is an essential factor in gaining
human trust in decision support systems (Wang &
Benbasat, 2008).
2.3 Shortcomings of Data Warehousing
and Big Data
A first drawback of the two concepts refers to the
associated methodologies, which do not lay enough
emphasis on understanding the fundamental
information needs of the decision maker rather than
rushing to data collection and interpretation (Seen &
Sinha, 2005).
In addition, there is minimal or inexistent
correlation between internal- and external data
sources (i.e., connecting the endogenous and the
exogenous information flows - see Figure 1).
Further on, insufficient effort is put into realising
what data is needed for being able to draw useful
inferences, but is unavailable. Even the recent
approaches proposing to limit the amount of sensor
data used in situation assessment (e.g. by switching
on- or off additional pre-stored sensor data sources)
is in fact relying on the command and control to
pinpoint what data should be taken into account to
possibly change the narrative.
If the above deficiency is identified, then the need
for data that is not available, but is deemed necessary,
may become the source of additional data collection
tasks; however, this can inadvertently result in poor
data quality ((Hazen, Boone, Ezell, & Jones-Farme,
2014; SAS, 2015)).
Another issue is the limited progress in
transforming existing building management
processes to produce the necessary data as a by-
product of the production (or service delivery)
process, instead of requiring additional data entry
(Hazen et al., 2014) (a main source of data quality
issues as shown above).
There has been a tendency to disregard the
collected data context (Corrigan, 2007), thus creating
the danger of situation mis-identification without
even being aware of having committed this mistake
(Santanilla, Zhang, Althouse, & Ayers, 2014).
Another important aspect is the typical reliance of
big data technology on machine learning techniques
producing models whose uncertainty cannot be
adequately assessed and whose predictions cannot be
adequately explained (Kendall & Gal, 2017).
To conclude, the main shortcomings found are:
- On each decision-making level, one must correlate
internal and external data.
- With the opportunity to collect and access very large
amounts of building–related data, the typically low
density of useful content (Li et al., 2017) makes it
difficult to identify patterns that are useful for
decision making (too many patterns identifiable by
algorithms) – unless one uses heuristics (i.e., the
result of prior learning) to discern what is relevant
(note that the measure of relevance may change in
time and with the current interpretation of data).
3 BUILDING INFORMATION
MODELS FOR EFFECTIVE
DECISIONS
3.1 The OODA Loop as an Activity
Network
The tasks that appear in each type and level of
decision-making and the feedback that can be used to
inform the filters used to selectively observe reality
might be studied using a model that explains how
successful decisions are made. This model is part of
the Observe, Orient, Decide and Act (OODA) Loop
devised by John Boyd (Osinga, 2006).
Although some authors such as Benson & Rotkoff
(2011) understand OODA to be a strict sequence of
tasks, this is in fact not true due to the feedback links
inside the high level ‘loop-like’ structure that are
responsible for learning and for decisions about the
kind of filters necessary. Thus, OODA is in fact an
activity network featuring rich information flows
among its activities and the environment.
Situation-aware Building Information Models for Next Generation Building Management Systems
77
A brief review of Boyd’s OODA concept can be
used to highlight potential development directions for
data warehousing and/or big data methodology for
decision support. Thus, decisions can be made by the
management / command & control system of a BMS,
in any domain of action and on any level or horizon
of management (i.e., strategic, tactical, operational
and real-time, performing four interdependent tasks.
The tasks that appear in each type and level of
decision-making and the feedback that can be used to
inform the filters used to selectively observe reality
may be studied using a model that explains how
successful decisions are made.
Note that this ‘loop’ is often misunderstood to be
a strict sequence of tasks (Benson & Rotkoff, 2011).
OODA is not a strict loop, due to the feedback links
inside the high level ‘loop-like’ structure that are
responsible for learning and for decisions about the
kind of filters necessary. Thus, in fact it is actually an
activity network featuring rich information flows
among the OODA activities and the environment.
This learning has the potential to result in
decisions that also emphasize relevant gaps and thus
initiate capability improvement efforts. Self-reflective
management typically engages in such learning,
comparing the behaviour of the external world and its
requirements on the system (the future predicted
action space) with the action space of the current
system (including the current system’s ability to
sense, orient, decide and act). Note that here, the term
‘action space’ describes the set of possible outcomes
reachable using the system’s current resources
(technical, human, information and financial).
The learning loop is in itself an OODA loop
analogous to the one discussed above, although the
ingredients are different and closely associated with
strategic management (see Figure 2). Thus the
OODA-style questions are in this case: a) what to
observe, b) how to orient to become situation-aware
and c) what is guiding the decision about what to do
(within constraints, decision variables and possible
actions) so as to be able to act. The action space of
this strategic loop consists of transformational actions
(company re-missioning, change of identity, business
model change, capability development, complete
metamorphosis, etc.).
Strategic self-reflection compares the current
capabilities of the system to desired future
capabilities, enabling management to decide whether
the change will affect the system’s capabilities
(including decision making capabilities), the system’s
identity (re-missioning), or both. Note that the
management may also decide to instead
decommission that part of the system due to its
inability to fully perform the system’s mission.
Such transformations are typically implemented
using a separate programme or project using a similar
suitable iterative paradigm, such as the so-called
Plan-Do-Check-Act (PDCA) loop (Lawson, 2006),
possibly in a recursive manner (Schmidt, Elezi,
Tommelein, & Lindemann, 2014) e.g. for complexity
control.
Figure 2: Extended OODA Loop as an activity network (based on (Fadok, Boyd, & Warden, 1995)) featuring additional
Learning and Narratives Loops.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
78
3.2 Consequences for BMS Decision
Support based on Data
Warehousing and Big Data –
Enabled BIM
The above analysis implies that ‘big data’ (meaning
the collective technologies and methods of data
analysis and predictive analytics) has the potential to
enable situational awareness (a condition of
successful action) by delivering a plethora of
previously unavailable domain-level facts and
patterns relevant for decision-making. However, this
data needs to be interpreted, which calls for a theory
of situations resulting in a narrative of what is being
identified or predicted. Without such a narrative,
there is no true situational awareness or trust in the
system, which can substantially limit the chances of
effective action.
It is therefore argued that having the ability to
gather, store and analyse large amounts of data using
only algorithms is not a guarantee that the patterns
thus found in data can be turned into useful and
trustworthy information that forms the basis of
effective decision-making, followed by appropriate
action leading to measurable success.
Importantly, the process is similar the other way
around: when interpreting available data, there can be
multiple fitting narratives; unfortunately, it is quite
difficult to choose the ‘correct’ one. In this case,
adequate means of reasoning with incomplete
information could help articulate a need for new data
(or new types of data) that can resolve the ambiguity.
As a result of the above reasoning, the authors
argue that supporting decision-making based on data
warehousing using ‘big data’ requires the collection
of a second level of data. This ‘second level’ is not
meant to refer to particular facts, but rather to
underpin the creation of an inventory of situation
types, containing facts that must be true, facts that
must be not true, as well as constraints and rules of
corresponding causes and effects. These situation
types can be considered models (or model prototypes)
of the domain, which can be matched against findings
on the observed data level.
Note that due to the changing nature of the
Universe of Discourse, the above-mentioned situation
types are also expected to evolve; therefore, one
should not aim to design and / or construct a facility
that relies on a completely predefined ontology of
situation types. Rather, there is a need for a capability
to continuously improve and extend this type of
knowledge, including the development and learning
of new types, which are not a specialisation of some
known type. This is required in order to ensure that
the ‘world of situations’ remains open, as described
by Goranson and Cardier (2013).
In order to achieve adequate situation awareness
for effective decision making, collected data needs to
be filtered based on relevance (Li et al., 2017;
Szafranski, 1995), dictated by the possible situations
of interest. However, as the current situation is
typically not unambiguously known and changes as
data gathered is interpreted, one will have to maintain
a dynamic narrative (or set thereof) of the situation,
which will continually adjust the data needs (Madden,
2012) as well as what needs to be filtered out, or be
kept. This constitutes yet another OODA loop,
applied to the set of narratives involved in the
interpretation of data for decision making (see
Narratives Loop in Figure 2).
4 DESIGN PRINCIPLES FOR
‘NEXT GENERATION’ BMS
DECISION MAKING
On both existing and emergent system levels,
decision making needs a timely and accurate
narrative that looks behind the ‘observables’. An
essential aspect of agile and effective decision
making (whether on strategic or tactical level) relies
on the ability of the system in question to create and
to continually update its situated insight, thus being
able to deal with uncertainty.
4.1 First Functional Principle: Employ
Situated Reasoning
Consider the following domain-level observations
within the domain of building energy demand at
cluster level and its coupling with complex integrated
energy systems; O1: some buildings within an
‘intelligent’ cluster report increased levels of energy
consumption; O2: occupants of some of the buildings
within the cluster (some overlap with O1) report
reduced thermal comfort, O3: there are no reports of
malfunction. Then, O4: energy consumption goes to
normal however O5: occupants still complain. Then
later on, the situation (O1 – O5) repeats itself.
A domain-level theory could describe energy
supply-demand rules in complex smart building
clusters including e.g. energy consumption
fluctuations depending on various factors.
While it is possible to build a model of the
situation, it is quite difficult and computationally
intensive to know all the intricacies (in space and
time) of such complex interconnected systems within
Situation-aware Building Information Models for Next Generation Building Management Systems
79
the building cluster (Frayssinet et al., 2018). More
importantly, the information usable to address the
situation with the right decision is incomplete.
Therefore, one does not really know what really
causes the increased energy consumption and thermal
discomfort in the absence of well-identified
malfunctions.
Typically, the Service Entity (SE) attempting to
address the situation must be able to interpret past
observed events and build types of situations (the
candidate interpretations of the events described),
each with their own logic and constraints. Thus, a
competent SE would be familiar with a repertoire of
situation types and their internal logic. The question
is: which one of these can be used as the correct
narrative of the observed events in this case?
The SE can interpret the current situation S based
on known matching situation types S
A
– e.g. sensor
measurement drift (e.g. due to low quality control or
bugs in embedded software) in some batches and S
B
,
e.g. semantic communication problem between
BMS-es within the cluster. Note that there may be a
myriad of situation types S
N
, some being variations of
a general type.
The SE’s Decision Suppot Sstem (DSS) can
employ situated reasoning in order to find out what
(perhaps very simple) additional fact/s would have to
be discovered to disambiguate between S
A
and S
B
and
thus be able to act appropriately.
While such avenues of action may seem rather simple
for humans, the real issue is the potential automation
and optimisation of such behaviour in BMS, based on
interpretation of BIM data. In order to achieve this,
one needs to employ a second functional princciple.
4.2 Second Functional Principle:
Channel Theory for Situation
Awareness
If situations are organised in types and their internal
logic is known (like in the scenario in Section 4.1),
then there exist possible ‘channels’ through which
information in one situation type can be transferred
(possibly in a lossy manner) to another situation type.
A recently popularised mathematical approach of
the above is the category theoretic treatment of
situation theory (Barwise & Perry, 1998; Devlin,
2003; Goranson & Cardier, 2013). The mechanism
that allows the two levels (situation theory and
domain level theory(ies)) to coexist is channel logic
(Barwise & Seligman, 2008) - according to which,
given the category of situations representing situation
types, there is a mapping that regulates the way
complete lines of reasoning can be ‘transplanted’
from one situation type to another.
This transplanting works as follows: when there
exists a logic in a known situation type S
A
and the
facts suggest that the situation is of a related type S
B
,
many (however typically not all) facts and inferences
should also be valid in type S
B
(see Figure 3).
As a result, if we have a known situation of type
S
A
with facts supporting this claim, and we only have
scarce data about another situation type of interest (of
type S
B
), channel logic allows us to deduce the need
for data that can be used to ‘fill in the details’ about
this second situation of type S
B
. The resulting
mapping is a so-called morphism between categories
and can be implemented using functional
programming techniques. In Figure 3, this (info-)
morphism is represented by the double-headed
arrows shown carrying information (while possibly
losing some, as previously stated) from one situation
type to another type to another.
In the case described in section 4.1, this morphism
may mean analysing particular sensor manufacturing
data and/or simulating events on the BMS in question
while observing the effects (as the original events
appear to be of a transient and random nature, either
due to random sensor drift or complex unwanted
interaction within BMS software or even sensor
embedded software).
The practical consequence is that the decision
maker can use this analogical reasoning to come to
valid conclusions in an otherwise inaccessible
domain; should this not be possible, it allows to at
least narrow down the need for specific data that can
support a valid conclusion.
The above also illustrates in a simplified way the
ability of the situation theoretic logic to infer that for
decision making, there is a need for specific, but yet
unavailable data that can disambiguate the
interpretation of what is known at the time.
5 CONCLUSIONS AND FURTHER
WORK
In the context of the increasing rate of change and the
resulting flood of data, decisions, even in traditionally
local and non-time-critical domains such as BMS will
have increasingly far-reaching consequences and
need to be promptly made - often, in real time.
The work presented in this paper can be used towards
creating an ongoing situational awareness capability.
The effective use of ‘proven repertoires’ nowadays
increasingly depends on their fast, near-automated
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
80
Figure 3: Simplification of Category Theoretic approach of Situation Theory (Noran & Bernus, 2018).
deployment, which is traditionally based on human
tacit skills and knowledge.
The paper has shown how novel paradigms can
assist traditional data analytics approaches in order to
achieve adequate BIM situation awareness and thus
properly support decision-making within complex
Building Management Systems-of-Systems.
Further work will continue to focus on the
principles underpinning situation theory-based
decision-making and related supporting technology.
The results will be used to demonstrate their use in
complex Building Management Systems-of-Systems
within clusters of Intelligent Buildings present in the
emerging Smart City paradigm.
REFERENCES
Adelman, S., & Moss, L. (2000). Data Warehouse Risks in
Data Warehouse Project Management: Addison WL
Barwise, J., & Perry, J. (1998). Situations and Attitudes:
Center for the Study of Language and Inf.
Barwise, J., & Seligman, J. (2008). Information Flow: The
Logic of Distributed Systems. In Cambridge Tracts in
Theoretical Computer Science. Cambridge: Cambridge
University Press.
Bayler, M. (2016). The Liquid Enterprise: How the network
is transforming value, what it means for business, and
what leadership needs to do about it. UK: Infinite Ideas.
Benson, K., & Rotkoff, S. (2011). Goodbye, OODA loop:
A complex world demands a different kind of decision-
making. Armed Forces Journal, 149(3), 26-28.
Corrigan, R. (2007). Digital Decision Making: Back to the
Future: Springer Verlag.
Devlin, K. J. (2003). Logic and Information. Cambridge:
Cambridge University Press.
Dibb, S., Meadows, M., & Wilson, D. (2015). Digitisation
and Decision Making in the Boardroom. New Economic
Models in the Digital Economy. Retrieved from
http://www.nemode.ac.uk/wp-
Doumeingts, G., Vallespir, B., & Chen, D. (2003). GRAI
Grid Decisional Modelling. In P. Bernus, K. Mertins, &
G. Schmidt (Eds.), Handbook on Architectures of
Information Systems (pp. 313-339). Heidelberg:
Springer Verlag.
Fadeyi, M. O. (2017). The role of building information
modeling (BIM) in delivering the sustainable building
value. International Journal of Sustainable Built
Environment 6, 711-722.
Fadok, D. S., Boyd, J., & Warden, J. (1995). Air Power’s
Quest for Strategic Paralysis: Air University Press.
Forte Asset Services. (2019). Building Management
System Facts and Terms. Retrieved from
https://www.forteassetservices.com.au/building-
management-systems-bms/facts-terms/
Frayssinet, L., Merlier, L., Kuznik, F., Hubert, J., Milliez,
M., & Roux, J. (2018). Modeling the heating and
cooling energy demand of urban buildings at city scale.
Renewable and Sustainable Energy Reviews, 81, 2318-
2327.
Gandomi, A., & Haider, M. (2015). Beyond the hype: Big
data concepts, methods, and analytics. International
Journal of Information Management, 35, 137-144.
Goranson, T., & Cardier, B. (2013). A two-sorted logic for
structurally modeling systems. Progress in Biophysics
and Molecular Biology, 113, 141-178.
Greenhalgh, T., Howick, J., & Maskrey, N. (2014).
Evidence based medicine: a movement in crisis? British
Medical Journal, 348, 3725.
Hazen, B. T., Boone, C. A., Ezell, J. D., & Jones-Farme, A.
D. q. f. d., predictive analytics, and bigdata in supply
chain management: An introduction to the problem and
suggestions for research and applications. Int J.
Production Econ 154:72-80. (2014). Data quality for
datascience, predictive analytics, and bigdata in supply
Facts
..
.
.
.
Inferences
..
.
.
.
Situation Type
S
A
Facts
..
.
Inferences
..
Need to recognise S as being of
type S
A
or S
B
Situation Type
S
B
Facts
..
.
Inferences
..
Situation S
Is of
type A?
Is of
type B?
Information Channel
Disambiguating
Information
Situation-aware Building Information Models for Next Generation Building Management Systems
81
chain management: An introduction to the problem and
suggestions for research and applications. Int J.
Production Econ, 154, 72-80.
Horitaa, F., de Albuquerquea, J., Marchezinid, V., &
Mendiondoc, E. (2016). Bridging the gap between
decision-making and big data sources: Application of
model-based framework to disaster management in
Brazil. Decision Support Systems, 97, 12-22.
Inmon, W. H., Zachman, J. A., & Geiger, G. J. (1997). Data
Stores, Data Warehousing and the Zachman
Framework. New York, NY 10011: McGraw-Hill.
Kendall, A., & Gal, Y. (2017). What Uncertainties Do We
Need in Bayesian Deep Learning for Computer Vision?
In 31
st
Conference on Neural Information Processing
Systems (NIPS 2017). Long Beach, CA.
Kimball, R. (1996). The Data Warehouse Toolkit. New
York, N.Y.: Wiley & Sons.
Kościelniaka, H., & Putoa, A. (2015). BIG DATA in
decision making processes of enterprises. Procedia
Computer Science, 65, 1052-1058.
Lawson, H. (2006). A journey through the systems
landscape. London: College Publications.
Li, M., Zhang, Z., Hu, Z., & 17., W. P. (2017). Big Data-
driven Technology Innovation: Concept and Key
Problems. In WHICEB 2017 Proceedings: AIS
Electronic Library.
Madden, S. (2012). From Databases to Big Data. IEEE
Internet Computing, 16(3), 4-6.
Marr, B. (2015). Big Data: Using Smart Big Data,
Analytics and Metrics to Make Better Decisions and
Improve Performance Wiley & Sons.
Matte, G., & Rizzi, S. (2009). A Survey on Temporal Data
Warehousing. Int Journal of data Warehousing &
Mining, 5(1), 1-17.
Noran, O., & Bernus, P. (2018). Improving Digital Decision
Making Through Situational Awareness. In B.
Anderssonm, B. Johansson, S. Carlsson, C. Barry, M.
Lang, H. Linger, & C. Schneider (Eds.), ISD: Designing
Digitalization. Lund, Sweden: Lund University.
Nowak, P., Ksiazek, M., Draps, M., & Zawistowski, J.
(2016). Decision Making using of Building Information
Modeling. Procedia Engineering, 153, 519-526.
Osinga, F. (2006). Science, strategy and war: The strategic
theory of John Boyd. London, UK: Routledge.
Pope, C. (2003). Resisting evidence: the study of evidence-
based medicine as a contemporary social movement.
Health, 7, 267-282.
Rokoeei, S. (2015). Building Information Modeling in
Project Management: Necessities, Challenges and
Outcomes. (Proceedings of 4th International
Conference on Leadership, Technology, Innovation and
Business Management), 87-95.
Santanilla, M., Zhang, D. W., Althouse, B. M., & Ayers, J.
W. (2014). What Can Digital Disease Detection Learn
from Google Flu Trends? American Journal of
Preventive Medicine, 47(3), 341-347.
Santos, E. T. (2009). Building Information Modelling and
Interoperability. Proceedings of the 13
th
Congress of the
Iberoamerican Society of Digital Graphics, Brazil.
SAS. (2015). From Data to Action: A Harvard Business
Review Insight Center Report. Retrieved from
http://www.idgconnect.com/
Satyanarayana, R. G., Srinivasu, R., Poorna Rao, M., &
Rikkula, S. R. (2010). Data Warehousing, Data Mining,
OLAP and OLTP Technologies Are Essential Elements
To Support Decision-Making Process In Industries. Int.
Journal on Comp. Sci. and Eng, 2(9), 2865-2873.
Schmidt, M. T., Elezi, F., Tommelein, I. D., & Lindemann,
U. (2014, 9-12 Dec. 2014). Towards recursive plan-do-
check-act cycles for continuous improvement. Paper
presented at the 2014 IEEE International Conference on
Industrial Engineering and Engineering Management.
Seen, A., & Sinha, A. P. (2005). Comparison of Data
Warehousing Methods. Comm ACM, 48(3), 79-84.
Szafranski, R. (1995). A Theory of Information Warfare.
Airpower Journal, 9(1), 1-11.
Teliceanu, M., Golovanov, N., Lazaroiu, G., & Dumbrava,
V. (2017). The importance of building clusters trading
in smart cities. International Conference on ENERGY
and ENVIRONMENT (CIEM).
Van Nederveen, G. A., & Tolman, F. P. (1992). Modelling
multiple views on buildings. Automation in
Construction, 1(3), 215-224.
Wang, W., & Benbasat, I. (2008). Attributions of Trust in
Decision Support Technologies: A Study of
Recommendation Agents for E-Commerce. Journal of
Management Information Systems, 24(4), 249-273.
doi:10.2753/MIS0742-1222240410
Wei, T., & Zhu, Q. (2016, 22-25 May 2016). Co-scheduling
of flexible energy loads in building clusters. Paper
presented at the 2016 IEEE International Symposium
on Circuits and Systems (ISCAS).
Zdravković, M., Noran, O., & Trajanović, M. (2014).
Towards Sensing Information Systems. In Proceedings
of 23
rd
International Conference on Information
Systems (ISD2014). Berlin: Springer Verlag.
Zhang, X., Lovati, M., Vigna, I., Widén, J., Han, M., Gal,
C., & Feng, T. (2018). A review of urban energy
systems at building cluster level incorporating
renewable-energy-source (RES) envelope solutions.
Applied Energy, 230, 1034-1056.
ICEIS 2020 - 22nd International Conference on Enterprise Information Systems
82