Friday schedule

Full Programme
Wednesday, OCT 11
Thursday, OCT 12
Friday, OCT 13

Friday, October 13th, 2023


Morning Coffee and Tea



Personalized Energy Saving Recommendations for Effective Conservation: Leveraging Data Analytics, Smart Energy Sensors, and an Energy App (Personalisierte Energiespar-Empfehlungen für wirksame Energieeinsparung: Nutzung von Daten-Analytik, intelligente Energiesensoren und eine Energie-App)
Predicting Particulate Matter in Poland using a CNN-LSTM Neural Network (Vorhersage von Feinstaub in Polen mit Hilfe eines neuronalen CNN-LSTM-Netzwerks )
Assessing the Impact of TLS-Derived Vegetation Structure on Microclimatic Variability in Taita Hills, Kenya (Einfluss dreidimensionaler Vegetationsstruktur auf die Variabilität des Mikroklimas in den Taita Hills, Kenia)
Student Prize Awards


Coffee and Tea Break


Main Track: Environmental Modeling Advances
GAEA - A Country-Scale Geospatial Environmental Modelling Tool: Towards a Digital Twin for Real Estate
Monitoring the physical and artificial environment at large-scale is crucial for approaching significant problems such as climate change, biodiversity loss, and sustainable urban growth. Towards this direction, GAEA is a novel AI-empowered geospatial online tool, designed to facilitate country-scale environmental monitoring, modelling, analytics, and geo-visualizations, providing valuable insights in the geographical region of the country of Cyprus, with some focus on the real estate application domain. This paper presents the design and development of GAEA, the needs and requirements it addresses, the environmental services it offers, its implementation details and main features, and an evaluation and discussion of its perspectives and overall potential. GAEA offers a user-friendly web interface that allows users to interact with a wide range of services, including land use monitoring, climate information, geohazard, and proximity analysis. GAEA is an important milestone and real-world demonstration of the vision of creating a country-scale environmental digital twin that allows informed decisions in land use assessment, climate analysis, and disaster management.
Digitalization of the Value Chain Pig Production - Discussion of Novel Approaches and Application of Self-Sovereign Identity
Livestock management is adapting to consumer demands with the aid of Precision Livestock Farming (PLF) and innovative technologies like blockchain and Self-Sovereign Identity (SSI). In this paper, we raise the question if SSI can be leveraged for creating decentralized digital identities, particularly in pig production, by discussing three proposals towards SSI adoption in the value chain of pig production. We discuss employing Sovrin’s thing controller approach, followed by a proposal for pig representation through Verifiable Credentials (VC) or dynamic Non-Fungible Tokens (NFT). Scalability (in terms of the number of wallets or the number of VCs) and ownership transfer (along with underlying transaction costs) emerge as critical challenges, while general feasibility is given from a high-level perspective. However, based on the potential for enhanced transparency and traceability, we argue to pursue further empirical research while highlighting a research direction towards decision support for choosing a proper SSI framework.
Towards Data Spaces for circular economy and green business value networks
Circular economy (CE) has been identified by several studies as the necessary reformation of the industry to decrease the environmental impact of production in the fight against climate change. Some studies have identified the lack of technological solutions to support the move towards a circular economy where among others the digital networking and data exchange is one of the most pressing and general problems which must be solved cross-industry and cross-country. This paper therefore identifies the most important requirements for a digital infrastructure to support CE and proposes a solution that combines all these factors by using Data Space concepts and technologies as the backbone for collaboratively collecting data in form of Digital Product Passports (DPP).
Detecting effects on soil moisture with Guerilla Sensing
A soil moisture monitoring system (SMOS) is presented to support the microclover project in determining effects of soil cover on soil moisture. It is built as a project-specific adaptation of the environmental information system (EIS) Guerilla Sensing. We describe the adaptation process step by step to provide a blueprint for easy use of Guerilla Sensing in similar future projects.
AI and Sustainability III
Detection of wind turbine motion between satellite bands with convolutional neural networks
Due to the design of the MutliSpectral Instrument on-board Sentinel-2 Satellite, each spectral band observes the ground surface at different times. Recent studies have examined this temporal band offset, mostly to predict aircraft or ship velocities but also for the movement of waves. However, these methods come with the premise of somewhat consistent backgrounds and clear horizontal movements. This paper presents an approach for detecting motion between blue, green and red satellite bands in changing environments to evaluate wind turbine motion using a convolutional neural network (CNN). With this technique, an automatic recognition of turbine activity has been developed, avoiding the barely visible vertical motion in the labelling stage by focusing on the shadow spin. This has been done by creating open source data that is used for training and validation. By focusing on a binary classification approach between spinning wind turbines and non-turbine images, it has been found that a classification is possible and accurate with this method. In addition, limitations and peculiarities of the data and the band offset are described, including an analysis of the occlusion sensitivity. This detection can be useful for precise remote sensing of activity at a given location and is therefore not only of interest to the wind energy industry, which currently only works with proprietary data for energy efficiency or other activity based turbine improvements, but also for environmental monitoring and protection.
Proof of concept for a new battery sorting method based on deep learning image classification
Battery recycling requires efficient sorting based on chemical composition. Traditional methods like X-Ray or Electromagnetic Sensors lack automation, with X-Ray sorting 26 batteries and electromagnetic sorting only 6 batteries per second. We propose using deep learning image classification to detect battery manufacturer and product series. Our prototype includes a conveyor belt, webcam, ring light, and Nvidia Jetson AGX Orin. With a dataset of 9 battery series, we achieved over 99% validation accuracy using a pretrained MobileNetV2 model. The model can classify 50 images per second with limited hardware. This approach offers potential for automated sorting, significantly improving recycling throughput and efficiency. Further research should expand the dataset and explore applicability to other battery types, optimizing the model and hardware configuration.
Graph-Based Time Token Recognition using Graph Neural Networks and Stanza Library
This publication aims to explore the capabilities of the Stanza NLP library on one hand and Graph Neural Networks on the other hand by combining them in a time token classification task. After providing information on the German ”WikiWarsDE” dataset that is used for training, the evaluation dataset consisting of environmental documents as well as the transformations applied to both datasets to enhance the model, the setup of the Neural Network is presented. Afterwards, some early test results are evaluated before possible enhancements to the model are suggested.
Development of a smart farming dashboard based of 5G mobile Data
This work in progress paper is written as a short description mainly of the backend of project 5G, which is in the field of smart farming. The project focuses on using different technologies and machines for weed management. This work in progress paper highlights the need for efficient weed management. It discusses the problems which are associated with weed management and it raises questions that need to be addressed in this domain. Moreover, the topic of using weed management 5G, UAV (unmanned aerial vehicle) and field robotics in agricultural and farming services is an important topic at present. Besides, the work in progress paper shows possible technical concepts and processes which can be implemented into smart farming to increase its efficiency. This paper discusses special methods, which can be used in weed management by using AI (artificial intelligence). In addition to the project description, the paper includes an evaluation of the current state of the research and an outlook of potential future research.
Green Coding III
Influence of Static Code Analysis on Energy Consumption of Software
This paper investigates how the implementation of suggestions from static program analysis, performed by Pylint, a so-called linter for the programming language Python, affects the energy consumption of software. For this purpose, the energy consumption for algorithms implemented in the Benchmarks Game [Gob] were measured before and after the revision and the results compared. Early findings suggest that resolving the issues presented by Pylint can have a negative impact on energy consumption. This was the case in 3 out of 8 algorithms. The remaining cases showed no significant difference. We assume that the increased energy consumption is due to the multitude of possibilities to implement proposals, leading to a possibility for worsening performance. Further research and experimentation is needed to objectively evaluate the impact of Pylint and static program analysis by extension, on energy consumption.
Tactics for Software Energy Efficiency: A Review
Over the years, software systems experienced a growing popularization. With it, the energy they consume witnessed an exponential growth, surpassing the one of the entire aviation sector. Energy efficiency tactics can be used to optimize software energy consumption. In this work, we aim at understanding the state of the art of energy efficient tactics, in terms of activities in the field, tactic properties, tactic evaluation rigor, and potential for industrial adoption. We leverage a systematic literature review based on a search query and two rounds of bi-directional snowballing. We identify 142 primary studies, reporting on 163 tactics, which we extract and analyze via a mix of qualitative and quantitative research methods. The research interest in the topic peaked in 2015 and then steadily declined. Tactics on source code static optimizations and application level dynamic monitoring are the most frequently studied. Industry involvement is limited. This potentially creates a vicious cycle in which practitioners cannot apply tactics due to low industrial relevance, and academic researchers struggle to increase the industrial relevance of their findings. Despite the energy consumed by software is a growing concern, the future of energy efficiency tactics research does not look bright. From our results emerges a call for action, the need for academic researchers and industrial practitioners to join forces for creating real impact.
Sustainable Software Engineering: Patterns and Trends through Artifacts from a Practitioner's Perspective
This study aims to uncover trends and patterns in sustainable software engineering research, with a particular focus on artifact-oriented outcomes and a practitioner’s perspective. Continuous research on the topic of environmentally sustainable software engineering practices is essential to mitigate the environmental impact of software products and advance software processes that promote sustainability. Despite recognizing the issue, many software industry practitioners struggle to identify sustainability requirements for software products during software development. Many are unaware of any applied process models or other software engineering artifacts to support sustainability in software engineering practices. This working paper intends to map practitioner-focused outcomes and academic research in sustainable software engineering. By adopting a practitioner’s perspective, we categorize 11 types of software engineering artifacts. These artifacts represent tangible research outcomes for software practitioners and help with systematically analyzing academic publications on sustainable software engineering between 2001-2022. The analysis is based on a three-stage literature screening process, out of which three intermediate datasets are analyzed. The study provides valuable insights into the trends and patterns of research output, emphasizing the significance of artifacts and acknowledging their contribution to the field. The aim is to promote sustainable software engineering by considering and mapping the perspectives of both academics and practitioners. Furthermore, it opens up opportunities for future research and development.
Measuring environmental impact: Lessons learned

Within the recently completed SoftAWERE project, completed by the SDIA & Öko Institut on behalf of the Umweltbundesamt, we set out to measure the environmental impact of open-source libraries and tools.

With our talk, we would like to share key lessons learned:

(1) challenges in measuring power usage & barriers
(2) ideas & approaches for addressing measurement limitations
(3) tools we developed to simplify measurement as part of CI/CD process

We will also share further research areas we’ve identified as well as action items for the software community.


Wrap up


Farewell Lunch