Analytics and Decision Support for Green IS and Sustainability Applications

Permanent URI for this collection

https://hdl.handle.net/10125/102479

Browse

Recent Submissions

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Addressing Inertia in Pro-Environmental Behavior through Nudges: A Review of Existing Literature and a Framework for Future Research
    (2023-01-03) Beermann, Vincent; Haskamp, Thomas; Marx, Carolin; Uebernickel, Falk
    To counteract global warming, individuals must adopt pro-environmental behaviors, but many prefer their established behaviors because of inertia. This paper analyzes how we can address the inertia that hinders pro-environmental behavior using digital nudges. Our structured literature review finds 19 out of 20 studies that show how decision information nudges like feedback overcome behavioral inertia. Most of the habitual patterns we identified could be attributed to private household behaviors like inefficient energy or water consumption. We contribute a framework for how the three dimensions of inertia—behavioral, cognitive, and affective inertia—are best each addressed using informational, structural, and assistance nudges, respectively.
  • Thumbnail Image
    Item
  • Thumbnail Image
    Item
    Artificial Intelligence: The Future of Sustainable Agriculture? A Research Agenda
    (2023-01-03) Witte, Jonas; Gao, Kevin; Zöll, Anne
    Global warming and the increasing food demand are problems of the current generation and require a change towards sustainable agriculture. In recent years, research in the field of artificial intelligence has made considerable progress. Thus, the use of artificial intelligence in agriculture can be a promising solution to ensure sufficient food supply on a global scale. To investigate the state-of-the-art in the use of artificial intelligence-based systems in agriculture, we provide a structured literature review. We show that research has been done in the field of irrigation and plant growth. In this regard, camera systems often provide images as training/input data for artificial intelligence-based systems. Finally, we provide a research agenda to pave the way for further research on the use of artificial intelligence in sustainable agriculture.
  • Thumbnail Image
    Item
    Image-Based Sorghum Head Counting When You Only Look Once
    (2023-01-03) Mosley, Lawrence; Pham, Hieu; Bansal, Yogesh; Hare, Eric; Mwanza, Charity
    Modern trends in digital agriculture have seen a shift towards artificial intelligence for crop quality assessment and yield estimation. In this work, we document how a parameter tuned single-shot object detection algorithm can be used to identify and count sorghum heads from aerial drone images. Our approach involves a novel exploratory analysis that identified key structural elements of the sorghum images and motivated the selection of parameter-tuned anchor boxes that contributed significantly to performance. These insights led to the development of a deep learning model that outperformed the baseline model and achieved an out-of-sample mean average precision of 0.95.
  • Thumbnail Image
    Item
    Sustainable Energy System Planning in Developing Countries: Facilitating Load Profile Generation in Energy System Simulations
    (2023-01-03) Hart, Maria C. G.; Eckhoff, Sarah; Breitner, Michael H.
    Successful energy system planning is dependent on detailed electricity demand information. Especially in developing countries, pre-generated load profiles are often unsuitable as appliance ownership and usage vary significantly across borders, between urban and rural areas, and on household and industry levels. Synthesizing load profiles is often hindered by the inaccessibility of tools due to cost barriers, global unavailability, or required technical knowledge. As currently, no easily accessible and usable tool is available during energy system planning in rural areas of developing countries, we incorporate the open-source load profile generator RAMP into our web-based energy system simulator NESSI4Dweb+ to provide an intuitive user interface. We conduct an applicability check with self-collected data from a guesthouse in Sri Lanka, analyzing the impact of load distribution and magnitude on the economic, environmental, and reliable energy supply, that validates the artifact's relevance and ability to empower local decision-makers.
  • Thumbnail Image
    Item
    One Save Per Day: How Mobile Technology Can Support Individuals to Adopt Pro-Environmental Behaviors
    (2023-01-03) Ochmann, Jessica; Lehrer, Christiane
    The pressing issue of climate change requires humanity to reduce its ecological footprint drastically. While policymakers and companies must ensure the availability of green options, individuals are requested to contribute to the reduction of carbon emissions substantially. However, even when individuals recognize the need for pro-environmental behaviors, they often have difficulty meeting their expectations. Mobile technology for sustainability has the potential to support them in overcoming this issue by providing the decisive impetus for environmentally friendly behavior. Drawing upon the affordance perspective, we conducted a longitudinal qualitative study with users of a mobile app that encourages individuals to take daily sustainable actions. We present the affordance strands made possible by the app's features and how they lead to environmentally-friendly behaviors. We could observe behavior change with the app’s features. We also identify enablers and obstacles to affordance actualization. Our study contributes to Green IS research at the individual level and provides practical implications for mobile technology providers.
  • Thumbnail Image
    Item
    Energy Efficiency of Training Neural Network Architectures: An Empirical Study
    (2023-01-03) Xu, Yinlena; Martínez-Fernández, Silverio; Martinez, Matias; Franch, Xavier
    The evaluation of Deep Learning (DL) models has traditionally focused on criteria such as accuracy, F1 score, and related measures. The increasing availability of high computational power environments allows the creation of deeper and more complex models. However, the computations needed to train such models entail a large carbon footprint. In this work, we study the relations between DL model architectures and their environmental impact in terms of energy consumed and CO2 emissions produced during training by means of an empirical study using Deep Convolutional Neural Networks. Concretely, we study: (i) the impact of the architecture and the location where the computations are hosted on the energy consumption and emissions produced; (ii) the trade-off between accuracy and energy efficiency; and (iii) the difference on the method of measurement of the energy consumed using software-based and hardware-based tools.