Prioritization of Identified Data Science Use Cases in Industrial Manufacturing via C-EDIF Scoring
While data science and artificial intelligence (AI) can be highly beneficial for industrial manufacturers, it is not yet readily usable. Therefore, putting it to good use requires to understand the domain challenges and identify opportunities for deploying AI. Our work aims at solving this task by proposing a generalized framework for (1) exploring companies for use cases and (2) prioritizing them via C-EDIF scoring. This novel approach allows to determine the business importance of any use case by considering the underlying evaluability, data situation, impact and infeasibility. Besides the theoretical framework, our work also provides real-world insights from applying C-EDIF scoring in an extensive use case exploration phase. These results stem from a strategic partnership between data scientists and a renowned pump manufacturing company, where we successfully identified and rated opportunities for AI.
- Published in:
International Conference on Data Science and Advanced Analytics - Type:
Inproceedings - Authors:
Fischer, Raphael; Pauly, Andreas; Wilking, Rahel; Kini, Anoop; Graurock, David - Year:
2023
Citation information
Fischer, Raphael; Pauly, Andreas; Wilking, Rahel; Kini, Anoop; Graurock, David: Prioritization of Identified Data Science Use Cases in Industrial Manufacturing via C-EDIF Scoring, International Conference on Data Science and Advanced Analytics, 2023, https://ieeexplore.ieee.org/document/10302632, Fischer.etal.2023a,
@Inproceedings{Fischer.etal.2023a,
author={Fischer, Raphael; Pauly, Andreas; Wilking, Rahel; Kini, Anoop; Graurock, David},
title={Prioritization of Identified Data Science Use Cases in Industrial Manufacturing via C-EDIF Scoring},
booktitle={International Conference on Data Science and Advanced Analytics},
url={https://ieeexplore.ieee.org/document/10302632},
year={2023},
abstract={While data science and artificial intelligence (AI) can be highly beneficial for industrial manufacturers, it is not yet readily usable. Therefore, putting it to good use requires to understand the domain challenges and identify opportunities for deploying AI. Our work aims at solving this task by proposing a generalized framework for (1) exploring companies for use cases and (2) prioritizing...}}