Fraunhofer-Institut für Angewandte Informationstechnik FIT

Filters

Reset filters

Settings
Now showing 1 - 10 of 245
No Thumbnail Available
Publication

Pathways to Developing Digital Capabilities within Entrepreneurial Initiatives in Pre-Digital Organizations

2022 , Keller, Robert , Ollig, Philipp , Rövekamp, Patrick

To enable new digital business models, pre-digital organizations launch entrepreneurial initiatives. However, in developing the required digital capabilities, pre-digital organizations often face challenges as they are marked by the ways they have historically established their organizational identity. Research on how pre-digital organizations can develop digital capabilities remains scarce. This study draws on a single case study to illustrate potential pathways for the development of digital capabilities. Two key characteristics are identified: the source of digital capability development and the set-up of the actors involved. The authors synthesize four possible pathway manifestations, discuss the dynamic nature of pathway combinations, and suggest that managing a portfolio of pathways may be crucial for pre-digital organizations. Therefore, the study contributes to a better understanding of digital transformation in pre-digital organizations. Furthermore, it provides guidance for practitioners to reflect on when deciding which pathways to follow.

View
No Thumbnail Available
Publication

Explainable Long-Term Building Energy Consumption Prediction using QLattice

2022 , Wenninger, Simon , Kaymakci, Can , Wiethe, Christian

The global building sector is responsible for nearly 40% of total carbon emissions, offering great potential to move closer to set climate goals. Energy performance certificates designed to increase the energy efficiency of buildings require accurate predictions of building energy performance. With significant advances in information and communication technology, data-driven methods have been introduced into building energy performance research demonstrating high computational efficiency and prediction performance. However, most studies focus on prediction performance without considering the potential of explainable artificial intelligence. To bridge this gap, the novel QLattice algorithm, designed to satisfy both aspects, is applied to a dataset of over 25,000 German residential buildings for predicting annual building energy performance. The prediction performance, computation time, and explainability of the QLattice is compared to the established machine learning algorithms artificial neural network, support vector regression, extreme gradient boosting, and multiple-linear regression in a case study, variable importance analyzed, and appropriate applications proposed. The results show quite strongly that the QLattice should be further considered in the research of energy performance certificates and may be a potential alternative to established machine learning algorithms for other prediction tasks in energy research.

View
No Thumbnail Available
Publication

Impact of socio-economic factors on local energetic retrofitting needs - A data analytics approach

2022 , Ahlrichs, Jakob , Wenninger, Simon , Wiethe, Christian , Häckel, Björn

Despite great efforts to increase energetic retrofitting rates in the residential building stock, greenhouse gas emissions are still too high to counteract climate change. One barrier is that policy measures are mostly national and do not address local differences. Even though there is plenty of research on instruments to overcome general barriers of energetic retrofitting, literature does not consider differences in local peculiarities. Thus, this paper aims to provide guidance for policy-makers by deriving evidence from over 19 million Energy Performance Certificates and socio-economic data from England, Scotland, and Wales. We find that building archetypes with their respective energetic retrofitting needs differ locally and that socio-economic factors show a strong correlation to the buildings' energy efficiency, with the correlation varying depending on different degrees of this condition. For example, factors associated to employment mainly affect buildings with lower energy efficiency whereas the impact on more efficient buildings is limited. The findings of this paper allow for tailoring local policy instruments to fit the local peculiarities. We obtain a list of the most important socio-economic factors influencing the regional energy efficiency. Further, for two exemplary factors, we illustrate how local policy instruments should consider local retrofitting needs and socio-economic factors.

View
No Thumbnail Available
Publication

DeepKneeExplainer: Explainable knee osteoarthritis diagnosis from radiographs and magnetic resonance imaging

2021 , Karim, Rezaul , Jiao, Jiao , Döhmen, Till , Cochez, Michael , Beyan, Oya , Rebholz-Schuhmann, Dietrich , Decker, Stefan

Osteoarthritis (OA) is a degenerative joint disease, which significantly affects middle-aged and elderly people. Although primarily identified via hyaline cartilage change based on medical images, technical bottlenecks like noise, artifacts, and modality impose an enormous challenge on high-precision, objective, and efficient early quantification of OA. Owing to recent advancements, approaches based on neural networks (DNNs) have shown outstanding success in this application domain. However, due to nested non-linear and complex structures, DNNs are mostly opaque and perceived as black-box methods, which raises numerous legal and ethical concerns. Moreover, these approaches do not have the ability to provide the reasoning behind diagnosis decisions in the way humans would do, which poses an additional risk in the clinical setting. In this paper, we propose a novel explainable method for knee OA diagnosis based on radiographs and magnetic resonance imaging (MRI), which we called DeepKneeExplainer. First, we comprehensively preprocess MRIs and radiographs through the deep-stacked transformation technique against possible noises and artifacts that could contain unseen images for domain generalization. Then, we extract the region of interests (ROIs) by employing U-Net architecture with ResNet backbone. To classify the cohorts, we train DenseNet and VGG architectures on the extracted ROIs. Finally, we highlight class-discriminating regions using gradient-guided class activation maps (Grad-CAM++) and layer-wise relevance propagation (LRP), followed by providing human-interpretable explanations of the predictions. Comprehensive experiments based on the multicenter osteoarthritis study (MOST) cohorts, our approach yields up to 91% classification accuracy, outperforming comparable state-of-the-art approaches. We hope that our results will encourage medical researchers and developers to adopt explainable methods and DNN-based analytic pipelines towards an increasing acceptance and adoption of AI-assisted applications in the clinical practice for improved knee OA diagnoses.

View
No Thumbnail Available
Publication

Welcome

2022 , Cauchard, Jessica R. , Jarke, Matthias , Oliver, Nuria

View
No Thumbnail Available
Publication

Sustainable behavior in motion: designing mobile eco-driving feedback information systems

2022 , Gimpel, Henner , Heger, Sebastian , Wöhl, Moritz

Emissions from road traffic contribute to climate change. One approach to reducing the carbon footprint is providing eco-driving feedback so that drivers adapt their driving style. Research about the impact of eco-feedback on energy consumption is the basis for designing a mobile eco-driving feedback information system that supports drivers in reducing fuel consumption. This work develops design knowledge from existing knowledge. Subsequently, we implement a prototypical instantiation based on the derived knowledge. Insights from a field study suggest that our design artifact allows most drivers to decrease fuel consumption by 4% on average. The paper's theoretical contribution is a set of design principles and an architecture of the proposed mobile eco-driving feedback information system. One recommendation is to provide normative feedback that compares drivers with each other. This feedback appears to encourage drivers to decrease their fuel consumption additionally. The design knowledge may support researchers and practitioners in implementing efficient eco-driving feedback information systems.

View
No Thumbnail Available
Publication

Investigating the co-creation of IT consulting service value: Empirical findings of a matched pair analysis

2022 , Oesterle, S. , Buchwald, A. , Urbach, Nils

Digitalization is increasingly and broadly impacting on companies throughout all industries. To cope with digital transformation, organizations need specific IT skills and often face a bottleneck between required and existing capabilities. Thus, organizations revert to support from IT consultants. However, such collaborations need to create value so as to make client organizations future-proof in the long term. We therefore need a better understanding of how value is created in IT consulting projects. We build on service-dominant (S-D) logic as the theory base and evaluate our structural model, which explains IT consulting service value based on 77 matched pairs of IT consulting projects using structural equation modeling. We provide empirical support for the assumptions of S-D logic in the IT consulting industry and reveal determinants that significantly contribute to the overall IT consulting service value. Our results contribute to the ongoing discourse in the S-D logic literature and provide meaningful insights for practice.

View
No Thumbnail Available
Publication

AI-based industrial full-service offerings: A model for payment structure selection considering predictive power

2022 , Häckel, Björn , Karnebogen, Philip , Ritter, Christian

Artificial Intelligence and servitization reshape the way that manufacturing companies derive value. Aiming to sustain competitive advantage and intensify customer loyalty, full-service providers offer the use of their products as a service to achieve continuous revenues. For this purpose, companies implement AI classification algorithms to enable high levels of service at controllable costs. However, traditional asset sellers who become service providers require previously atypical payment structures, as classic payment methods involving a one-time fee for production costs and profit margins are unsuitable. In addition, a low predictive power of the implemented classification algorithm can lead to misclassifications, which diminish the achievable level of service and the intended net present value of the resultant service. While previous works focus solely on the costs of such misclassifications, our decision model highlights implications for payment structures, service levels, and - ultimately - the net present value of such data-driven service offerings. Our research suggests that predictive power can be a major factor in selecting a suitable payment structure and the overall design of service level agreements. Therefore, we compare common payment structures for data-driven services and investigate their relationship to predictive power. We develop our model using a design science methodology and iteratively evaluate our results using a four-step approach that includes interviews with industry experts and the application of our model to a real-world use case. In summary, our research extends the existing knowledge of servitization and data-driven services in the manufacturing industry through a quantitative decision model.

View
No Thumbnail Available
Publication

Steuerentlastung von kleinen und mittleren Einkommen. Umsetzungsmöglichkeiten im Rahmen der Tarifformel

2022 , Stöwhase, Sven , Teuber, Martin

View
No Thumbnail Available
Publication

Kindergarten for Free?! Empirical Evidence on the Utilization of Income Tax Deductions for Child Care Expenses

2021 , Calahorrano, Lena , Stöwhase, Sven

In most developed countries there exist ample possibilities for individuals to minimize their personal income tax burden by means of tax deductions. The associated revenue losses for the government have been estimated for a variety of tax deductions. However, relatively little is known about the share of eligible taxpayers who actually use these deductions, and, more specifically, about what determines utilization. The present paper tries to shed some light on this question in the context of the tax deductibility of expenses for child care in Germany. Using survey data on actual child care expenses and official tax-return data on deductions for child-care expenses, we derive utilization rates. We also analyze the determinants of utilization among those who filed a tax-return, using a subsample of the tax-return data. Our estimation results show that (potential) tax breaks from utilization are significantly positively correlated with the probability of utilization. Other kinds of deductions are also highly significant, suggesting that knowledge of tax statutes as well as opportunity costs matter. Moreover, we simulate the effects of a policy reform that enhances the generosity of deductions on the utilization rate. Such a reform would substantially increase utilization. Our results indicate that responses in utilization are more important than potential responses in labor supply.

View