Digital Twin in Industrial Application – Requirements to a Comprehensive Data Model
behavior. As the analysis of asset behavior is
mostly based on time series data, the config-
uration history of an asset's component
structure is of great relevance (e.g. for audit
purposes).
model if a fully-fledged basis for IIoT applica-
tions is to be achieved.
Behavioral Data
Data describing the behavior and operation
of assets takes the form of values from sen-
sors (such as temperatures, flows) and mes-
sages from controllers (e.g. error warnings,
ready states). Both types of data may appear
in different technical protocols and may
have significantly different characteristics:
Sensor data usually takes the form of a con-
tinuous stream of values while messages are
discrete and intermittent.
Environment
Assets are influenced by their environment
in terms of surrounding systems (e.g. car-
riages), installation sites (such as drilling
platforms) and environmental conditions
(such as hot or cold weather, presence of
dust, etc.). Thus, the proper recording of
such environmental data is a relevant as-
pect.
Environmental Data
Models and descriptions
Data describing the situation and the envi-
ronment in which the asset operates (e.g.
temperature, humidity) may originate from
sensors in the asset itself or its environment
or can be provided by web services. In light
of this variety of possible data sources, it is
obvious that different technical protocols
may be used to transfer this data (possibly
even in the same way as for the behavior
data). It therefore seems reasonable to con-
sider this data separately in a digital twin
data model.
A simple verbal description coupled with the
component structure may not suffice for all
the data analytics requirements relating to
an asset. Additional aspects such as geome-
try or the operating or maintenance situa-
tion (e.g. motor within a vehicle) are an es-
sential part of the digital twin data model.
Control Parameters (including history)
To define the operating profile of an asset,
the software programs and parameters used
(such as settings for motor speed, pump
pressures) must be recorded and stored.
Once more, the data history is of great rele-
vance because changing these parameters
modifies the behavior of the asset.
Finally, another type of data needs to be in-
cluded within a digital twin data model:
Connectivity Parameters
To be able to receive data from and send
data to an asset, information on how to ad-
dress the asset in the (inter)net is required.
This means unique addresses such as IP’s or
MAC declarations are required together with
a description of the authorization method.
In our opinion, the definition of a digital twin
data model is not complete without these
reference characteristics. As a smart asset
delivers data in an individual format, the de-
scription of these data streams (not the field
data in terms of “payload” itself but the
meta-information) must be part of the data
IIC Journal of Innovation
- 101 -