=Paper=
{{Paper
|id=None
|storemode=property
|title=Reactive Processes
|pdfUrl=https://ceur-ws.org/Vol-929/paper13.pdf
|volume=Vol-929
|dblpUrl=https://dblp.org/rec/conf/semweb/SolankiC12a
}}
==Reactive Processes==
https://ceur-ws.org/Vol-929/paper13.pdf
Reactive Processes
Monika Solanki∗ and Craig Chapman
Knowledge Based Engineering Lab
Birmingham City University, UK
{monika.solanki, craig.chapman}@bcu.ac.uk
Abstract. In this paper we present the Reactor Pattern to enable the
modelling of processes that consume inputs and produce outputs un-
der specific environmental conditions and on being triggered by certain
events. Reactor pattern is a content ontology design pattern and is espe-
cially targeted towards modelling reactive processes with a “black box”
view of the process.
1 Introduction
Many scenarios in the engineering, manufacturing and biotechnologies sectors
employ “reactive” processes, usually carried out in a closed system, e.g., a biore-
actor in which a chemical process is carried out, which involves organisms or
biochemically active substances derived from such organisms. Such processes
consume inputs and produce outputs in a controlled environment and on being
triggered by certain events. The purpose of the reactor pattern is to enable the
ontological modelling of such reactive processes in a generic way across multiple
domains. The reactor pattern is a content design pattern and provides ontologi-
cal placeholders for input and output parameters, environmental conditions and
events. The pattern exploits other CPs for the definition of certain entities.
2 Reactor Pattern
2.1 Intent
The remit of the reactor pattern is to enable the modelling of processes that
are reactive, consume inputs and produce outputs under specific environmental
conditions a.k.a. constraints and on being triggered by certain events. Reactive
processes are parametric where the governing parameters are process inputs and
outputs. The pattern can be instantiated to provide a knowledge level solution
to the problem of capturing parametric process related information in a domain
independent way.
∗
Principal and corresponding author
2.2 Competency Questions:
– What are the “types” of inputs consumed by a certain process?
– What are the “types” of outputs produced by a certain process?
– What are the values of parameters for a certain process?
– What is the measurement criteria for a specific parameter?
– What environmental conditions need to hold for the process to get activated?
– Which event triggers a specific process?
2.3 Some Conceptual Elements
– Process: placeholder for a process. The concept covers the definition of a
generic process.
– ParametricProcess: placeholder for a process governed by parameters.
– ReactiveProcess: a process specialising from ParametricProcess. Note
that in our abstraction, a reactive process needs to explicitly define at least
one input and output.
– ProcessParameter: an overarching entity representing parameters consumed
and produced by the process. The concept extends from Parameter defined
in the Parameter1 CP.
– InputParameter: a specialisation of the ProcessParameter representing the
input parameter. A process can consume several inputs.
– OutputParameter: a specialisation of the ProcessParameter representing
the output parameter. A process can produce several outputs.
– EmvironmentalCondition: an entity representing environmental conditions
governing the activation of the process. The condition may be specified as a
SWRL rule2 . There can be several environmental conditions for a process.
– Event: an event that triggers the process.
– hasEnvironmentalCondition: a relation between the environmental condi-
tion and the process.
– triggeredBy: a relation between the process and the environment.
– definesCondition: a relation between, EnvironmentalCondition and one
or more conditions it enforces.
It is worth noting that we explicitly abstract from providing further details on
how the environmental conditions are represented or how the parameter mea-
surements are defined. These are not part of the pattern definition. Well defined
existing vocabularies such as the LODE ontology3 for modelling events and the
QUDT vocabulary4 for measurement units should be exploited to provide defi-
nitions for these concepts.
1
http://www.ontologydesignpatterns.org/cp/owl/parameter.owl
2
http://www.w3.org/Submission/SWRL/
3
http://linkedevents.org/ontology/
4
http://qudt.org/1.1/vocab/dimensionalunit
2.4 Pattern Representation
The core concept in the pattern is a reactive process, parameterised with inputs
and outputs. A Manchester syntax rendering of the concept is illustrated below.
Class: ReactiveProcess
EquivalentTo:
ParametericProcess
and (hasInputParameter some InputParameter)
and (hasOutputParameter some OutputParameter)
and (triggeredBy some Event)
SubClassOf:
hasEnvironemntalCondition min 0 EnvironmentalCondition,
ParametericProcess
Figure 1 depicts the graphical representation of the reactor pattern 5 6 .
Fig. 1. Graphical Representation of Reactor Pattern
2.5 Consequences
The main advantage of this pattern is that its provides ontological modelling
capabilities for the inputs, outputs and environmental conditions that govern
reactive processes across several domains, independent of modelling details of
the actual reactor involved. This effectively caters for exposing a black box view
of the process, which is very desirable when querying the model for consumption
and production logistics of the process.
5
The OWL ontology for the pattern is available at
http://purl.org/biomass/ReactorPattern
6
Graphical representations of the pattern in this paper have been produced using a
trial version of the Maestro edition of TopBraid Composer.
2.6 Example usage: Algal Biomass Domain
As an exemplifier for the reactor pattern, we present a use case from the domain
of algal biomass. The set of inputs to the process of algal biomass cultivation are
carbon, water, total infrastructure area, total energy, nutrients, consumables and
labour. Possible outputs from the process are algal constituent products, indirect
algal products, uncaptured gas emission, liquid waste output, solid waste output.
Some environmental conditions that must hold for the algae to be harvested are,
– The water must be in a temperature range that will support the specific
algal species being grown.
– The pH range for most cultured algal species should be between 7 and 9,
with the optimum range being 8.2-8.7.
The event that triggers of the algae cultivation is the addition of the source
culture to the growing containers or reactors. Figure 2 depicts the application
of the reactor pattern.
Fig. 2. Graphical Representation of reactor pattern for modelling the algal cul-
tivation process
3 Summary
The reactor pattern provides a building block for the ontological modelling of
reactive processes. The pattern can be used across domains in scenarios where a
reactor is used to run processes that consume inputs to produce outputs under
controlled environmental conditions and when triggered by certain events. As
an example, the pattern has been applied to the algal biomass domain to model
the reactive process of algae cultivation.