Information Technology Reference
In-Depth Information
theory for adaptive systems that allows to reason not only about the “what” of moni-
toring and adaptation but also about the “how”. Finally, it claims that current
solutions are too much focused on isolated problems, and that we need an integrated
autonomic systems engineering approach to avoid undesirable feature interaction.
Within the software engineering domain, runtime adaptation has become a highly
relevant research topic. In this context, Model-Driven development and Software
Product Lines modeling techniques are used and extended to Dynamic Software
Product Lines that include variability transformations [3, 26, 8]. This work confirms
the research gap identified by [4] that the “how” of the adaptation is no longer some-
thing that can be abstracted from.
Another line of research is considering modeling and monitoring requirements for
adaptive systems. Goal-based approaches, such as Tropos [2] have been applied to the
specification of requirements of self-adaptive systems. Goal-based models are well
suited to exploring high-level requirements and it is natural to use goal models to
represent alternative behaviors that are possible when the environment changes. Goal-
based models have a natural fit with agent-oriented implementation platforms [11].
In the management literature, system theory and control theory have been around
for a long time. In the standard management control textbook of Simons [19], four
types of “control systems” are distinguished: diagnostic control systems, interactive
control systems, belief systems and boundary systems.
Diagnostic control
systems
correspond to the traditional cybernetic approach and are aimed at controlling results
using a closed control cycle. This mode of control is important but it also has its limi-
tations, according to Simons.
Belief systems
express norms and core values in the
organization that are aimed at controlling (or influencing) the value systems of the
people involved.
Boundary systems
constrain the behavior of the organization in
the face of risks to be avoided.
Interactive control
is focused on handling uncertain-
ties and on “doing the right thing”, rather than doing the things right, as in diagnostic
control, and is typically realized in the form of interaction.
3 A Framework for Management Service Design
3.1 The REA Business Ontology
The Resource-Event-Agent (REA) ontology was first formulated in [9] and has been
developed further, e.g. in [21,5,6]. The following is a short overview of the core con-
cepts of the REA ontology based on [23].
A
resource
is any object that is under the control of an agent and regarded as valu-
able by some agent. This includes goods and services. Resources are modified or
exchanged in processes. A
conversion process
uses some input resources to produce
new or modify existing resources, like in manufacturing. An
exchange process
occurs
as two agents exchange (provide, receive) resources. To acquire a resource an agent
has to give up some other resource. An
agent
is an individual or organization capable
of having control over economic resources, and transferring or receiving the control to
or from other agents. The constituents of processes are called
economic events
. REA
recognizes two kinds of duality axioms between events: conversion duality and
exchange duality.
