Delft University of Technology
Three Categories of Context-Aware Systems
Shishkov, Boris; Larsen, John Bruntse; Warnier, Martijn; Janssen, Marijn
DOI
10.1007/978-3-319-94214-8_12
Publication date
2018
Document Version
Final published version
Published in
Proceedings of Business Modeling and Software Design - 8th International Symposium, BMSD 2018
Citation (APA)
Shishkov, B., Larsen, J. B., Warnier, M., & Janssen, M. (2018). Three Categories of Context-Aware
Systems. In
Proceedings of Business Modeling and Software Design - 8th International Symposium, BMSD
2018
(Vol. 319, pp. 185-202). (Lecture Notes in Business Information Processing; Vol. 319). Springer.
https://doi.org/10.1007/978-3-319-94214-8_12
Important note
To cite this publication, please use the final published version (if applicable).
Please check the document version above.
Copyright
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent
of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Takedown policy
Please contact us and provide details if you believe this document breaches copyrights.
We will remove access to the work immediately and investigate your claim.
This work is downloaded from Delft University of Technology.
For technical reasons the number of authors shown on this cover page is limited to a maximum of 10.
Green Open Access added to TU Delft Institutional Repository
'You share, we take care!' - Taverne project
https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher
is the copyright holder of this work and the author uses the
Dutch legislation to make this work public.
Three Categories of Context-Aware Systems
Boris Shishkov
1,4(&)
, John Bruntse Larsen
2
, Martijn Warnier
3
,
and Marijn Janssen
3
1
Institute of Mathematics and Informatics,
Bulgarian Academy of Sciences, Sofia, Bulgaria
b.b.shishkov@iicrest.org
2
Department of Applied Mathematics and Computer Science,
Technical University of Denmark, Lyngby, Denmark
jobla@dtu.dk
3
Faculty of Technology, Policy, and Management,
Delft University of Technology, Delft, The Netherlands
{M.E.Warnier,M.F.W.H.A.Janssen}@tudelft.nl
4
Institute IICREST, Sofia, Bulgaria
Abstract. With regard to context-aware systems: some optimize system-
internal processes, based on the context state at hand; others maximize the
user-perceived effectiveness of delivered services, by providing different service
variants depending on the situation of the user; still others are about offering
value-sensitivity when the society demands so. Even though those three per-
spectives cover a broad range of currently relevant applications there are no
widely accepted and commonly used corresponding concepts and terms. This is
an obstacle to broadly understand, effectively integrate, and adequately assess
such systems. We address this problem, by considering a (component-based)
methodological derivation of technical (software) specifications based on
underlying enterprise models. That is because context states are about the
enterprise environment of a (software) system while the delivery of
context-aware services is about technical (software) functionalities; hence, we
need a perspective on both. We consider the SDBC (Software Derived from
Business Components) approach that brings together enterprise modeling and
software specification. On that basis: (a) We deliver a base context-awareness
conceptualization; (b) We partially align it to agent technology because adapting
behaviors to environments assumes some kind of pro-activity that is only fully
covered by agent systems, in our view. We partially illustrate our proposed
conceptualization and particularly - the agent technology implications, by means
of a case example featuring land border security.
Keywords: Modeling
System design Context-awareness SDBC
AORTA
1 Introduction
Whenever a group of entities collecti vely realize a goal, we consider them to belong to a
system [17]. An adapt able system has the ability to adjust to new conditions [8].
We argue that an essential feature of adaptable systems is context-awareness
© Springer International Publishing AG, part of Springer Nature 2018
B. Shishkov (Ed.): BMSD 2018, LNBIP 319, pp. 185–202, 2018.
https://doi.org/10.1007/978-3-319-94214-8_12
[22] – this is adjusting the system behavior depending on the situation at hand (context
state). Our observation is that context-aware systems are all about adjusting
“something” to the context state; however, what is adjusted differs: (i) Some
context-aware systems
optimize system-internal processes based on the context state at
hand [3, 12], for example: regulating the electro-consumption of home appliances for the
sake of keeping the overall building consumption within some boundaries. (ii) Other
context-aware systems
maximize the user-perceived effectiveness of delivered services,
by providing different service variants depending on the situation of the user [1], for
example: treating a distantly monitored patient in one way when his/her condition is
normal and in another way in case of emergency. (iii) Still other context-aware systems
are about
offering value-sensitivity when the society demands so [7 ], for example: in the
case of supporting judiciary processes, different levels of transparency are to be provided
to different categories of stakeholders . Not claiming exhaustiveness, we argue that those
three context-awareness perspectives cover a broad range of currently relevant applica-
tions, while corresponding system categorizations are observed to be missing in litera-
ture, especially as it concerns real-life (business) processes. Hence, we propose and
elaborate (in the following section) the above categorization.
Further, there are no widely accepted and commonly used concepts and terms with
respect to (i), (ii), and (iii). This is considered to be an obstacle with regard to broadly
understanding, effectively integrating,andadequately assessing such systems. We
address this problem, by taking a (component-based-) technical/software design per-
spective, assuming a desired methodological derivation of the technical (soft-
ware) specifications on the basis of underlying enter prise models. That is
because con text states are about the enterprise environment of a (software) system
while the delivery of context-aware services is about technical (software) functional-
ities; hence, we need a perspective on both. Stepping on previous work, we consider the
SDBC (
Software Derived from Business Components) approach that brings
together enterprise modeling and software specification [17, 18, 20]. On that basis:
• We deliver a base
context-awareness conceptualization (inspired by the cu rrent
discussion and the analysis carried out in Sect. 2) that is claimed to hold for all (i),
(ii), and (iii).
• We partially
align that concept ualization to agent technology [11, 27] because
adapting behaviors to environments is considered to assume some kind of
pro-activity that is only fully covered by agent systems, in our view.
We partially illustrate our proposed conceptualization and particularly - the agent
technology implications, by means of a case example featuring land border
security.
The remaining of the curren t paper is organized as follows: Sect. 2 elaborates and
discusses the above-mentioned context-awareness perspectives (optimizing internal
processes, maximizing the user-perceived effectiveness, being value-sensitive). In
Sect. 3, we: (a) briefly outline the SDBC approach that gives the general method-
ological guidelines we follow; (b) elicit (in line with SDBC) the base context-
awareness system concepts, referring to the analysis in Sect. 2; (c) enrich those
concepts from the AORTA (agent-technology) perspective; (d) distill on that basis a
meta-model that is considered a key contribution of this paper and propose guidelines
186 B. Shishkov et al.
refinements. Assuming the SDBC methodological guidelines and referring to a real-life
case study that is featuring the usage of drones in support of land border security, we
partially illustrate in Sect. 4 the meta-model and the AORTA framework influence,
emphasizing the applicability with regard to all three categories of context-aware
systems, as discussed above. We justify the adequacy of the delivered contribution in
Sect. 5, by analyzing related work. Finally, in Sect. 6 we conclude the paper.
2 System Behavior Perspectives
Referring to the notions considered in the Introduction, we will firstly elaborate on the
context-driven optimization of system-internal processes (Subsec t. 2.1), secondly – on
the contex t-driven maximization of the user-perceived effectiveness (Subsect. 2.2), and
finally, on the context-driven value-sensitivity (Subsect. 2.3). It is often that the
context-awareness is enabled by sensor technology [21] allowing to “know” what is
happening around; alternatively, there should be other ways of “sensing” the envi-
ronment [1]. As it concerns (i), (ii), and (iii) - see the previous section - this counts for
all of them. Further, considering the essence of their underlying system behaviors, we
use the following labels: SELF-MANAGING CONTEXT-AWARE SYSTEM for (i);
USER-DRIVEN CONTEXT-AWARE SYSTEM for (ii), and VALUE-SENSITIVE
CONTEXT-AWARE SYSTEM for (iii). Finally, even though most often context-
aware systems are “sensitive” to changes in the system environment, it is also possible
that the “sensitivity” is towards internal issues (things happening inside the system (not
in the environment) may trigger either internal optimizations, or changes in the services
delivered to the user, or a reconsideration of the “covered” values ). In this paper, we are
not restrictive with regard to the “sensitivity” (whether it concerns the system or the
environment).
2.1 Self-Managing Context-Aware System s (SMCAS)
SMCAS’ context-awareness is directed towards internal (system) optimiza-
tion purposes [16]. Such autonomic solutions [12] are proposed as a way to
reduce the cost of maintaining complex syst ems, and to increase the human a bility to
manage these systems properly, by automating (part of) their working. In essence,
self-managing system can be characterized by a feedback loop mechanism
that allows them to optimize their working based on input from the environment. For
the basic feedba ck loop, the system receives input from the environment (
monitor) and
can change its behavior which in turn has an effect on the environment (
effector). In
Autonomic Computing [10] this basic loop is extended into four components,
resulting in the MAPE Cycle, see Fig. 1 (left). Next to the “monitor” and “effector”
phases, the syst em internally has an
analyze phase that processes environmental input
and a
plan phase that changes the internal and external working of the system.
Taking this one step further, an autonomic system can manage another system
(the managed system) by placing it in the MAPE Cycle as shown in Fig. 1 (right).
Placed in such a configuration, the autonomic and managed systems together form a
SELF-MANAGING SYSTEM or self-adaptive system [14].
Three Categories of Context-Aware Systems 187
![Fig. 2. The UDCAS vision (inspired by [18, 22]): a schematic representation (left); Contextstates-driven system behavior variants (right)](/figures/fig-2-the-udcas-vision-inspired-by-18-22-a-schematic-2nwkr2u4.webp)
![Fig. 3. SDBC - outline (Source: [20], p. 48)](/figures/fig-3-sdbc-outline-source-20-p-48-1bos2km1.webp)
