![](bg8.png)
enable a good extensibility to a changing manufactur-
ing environment. Further, ontologies allow to trans-
fer our approach to other areas by adapting checking
rules and domain knowledge.
Another aspect to making software more sustain-
able is to use an event-based approach instead of pull-
based. We adress this by using a publish-subscribe
architecture (message broker). Further, the broker al-
lows to connect multiple instances of our GreenCC
and thus to balance workload. Since we developed
our GreenCC unit wise, it is further conceivable to
run each module on a different node.
Geiger et al. (Geiger et al., 2021) also mention
that software should be implemented in a lean way
and only perform exactly one task, what our system
complies with. Further, it is recommended to choose
the programming language wisely. Selection criteria
strongly depend on the specific use case. By using
Python, we are in the lower middle in terms of ex-
ecution time, energy and memory consumption (cf.
(Pereira et al., 2017)). In our evaluation, the focus is
on comparing different methods to validate manufac-
turing data. For better comparability, especially with
regard to the overheads through monitoring software,
all systems are implemented in the same language
and monitored with the same tools. In general, when
looking at our systems, we can see that the chosen
method already offers a decisive advantage in terms of
costs, energy consumption, and emissions. With fur-
ther regard to the integration in a globally operating
company and the resulting need for manageability of
the used language, Python offers an advantage at this
point compared to, for example, C or Go. However, it
is conceivable to implement parts of the system (e.g.
the stream handling unit) efficiently in future work.
6 CONCLUSION
We present a system for validating stream data in a
resource-efficient manner. Our GreenCC is a Python
based system that monitors incoming messages and
predicts inconsistencies based on patterns that occur.
For detailed analyses, a full consistency check can be
initiated. Our analyses have shown that compared to
our previous systems, energy consumption can be re-
duced significantly, especially when applying the sys-
tem to large manufacturing plants. The lower energy
consumption stands out in particular when consider-
ing CO
2
e emissions. Seen over the year, these can be
reduced in a medium-sized plant in the EU by a fac-
tor of about 0.6, which corresponds to 262 kgCO
2
e
(LighCC vs. Flink (all)). Overall, the use of our
GreenCC is profitable in each of our scenarios.
As the relevance for smart sustainable software is
existent, future work will continue to focus on green
data validation in manufacturing environments. Pat-
tern detection, as used in our GreenCC, offers many
opportunities, e.g. by using machine learning (ML)
algorithms. Depending on how resource consuming a
training process is and how often we have to re-train,
ML can offer a benefit when applying in a large het-
erogenous manufacturing environment.
REFERENCES
Cort
´
es, R., Bonnaire, X., Marin, O., and Sens, P. (2015).
Stream processing of healthcare sensor data: studying
user traces to identify challenges from a big data per-
spective. Procedia Computer Science, 52:1004–1009.
Gao, S., Dell Aglio, D., Pan, J. Z., and Bernstein, A. (2018).
Distributed stream consistency checking. In Interna-
tional Conference on Web Engineering, pages 387–
403. Springer.
Geiger, L., Hopf, T., Loring, J., Renner, M., Rudolph,
J., Scharf, A., Schmidt, M., and Termer, F. (2021).
Ressourceneffiziente programmierung.
Haav, H.-M., Maigre, R., Lupeikiene, A., Vasilecas, O., and
Dzemyda, G. (2019). A semantic model for product
configuration in timber industry. In Databases and
Information Systems X, volume 315, pages 143–158.
IOS Press.
Hendzlik, M., Lange, M., Kl
¨
ockner, P., Lambrecht, M.,
Frey, K., Dziekan, K., Dross, M., and Schmied, M.
(2022). Bausteine f
¨
ur einen klimagerechten verkehr.
Internationales Verkehrswesen, 74(1).
Paasche, S. and Groppe, S. (2022). Enhancing data qual-
ity and process optimization for smart manufacturing
lines in industry 4.0 scenarios. In Proceedings of The
International Workshop on Big Data in Emergent Dis-
tributed Environments, BiDEDE ’22, New York, NY,
USA. Association for Computing Machinery.
Pereira, R., Couto, M., Ribeiro, F., Rua, R., Cunha, J., Fer-
nandes, J. P., and Saraiva, J. (2017). Energy efficiency
across programming languages: how do energy, time,
and memory relate? In Proceedings of the 10th ACM
SIGPLAN international conference on software lan-
guage engineering, pages 256–267.
Tao, F., Qi, Q., Liu, A., and Kusiak, A. (2018). Data-driven
smart manufacturing. Journal of Manufacturing Sys-
tems, 48:157–169.
Tian, Y., Michiardi, P., and Vukoli
´
c, M. (2017). Bleach:
A distributed stream data cleaning system. In 2017
IEEE International Congress on Big Data (BigData
Congress), pages 113–120. IEEE.
Xuanyuan, S., Li, Y., Patil, L., and Jiang, Z. (2016). Config-
uration semantics representation: A rule-based ontol-
ogy for product configuration. In 2016 SAI Computing
Conference (SAI), pages 734–741. IEEE.
DATA 2023 - 12th International Conference on Data Science, Technology and Applications
628