PACIS 2022 Proceedings
Harnessing the Business Potential of Self-Service Machine Learning for Forecasting Warranty Costs – Insights from a Case Study in the Automotive Sector
New machine learning (ML) methods, such as AutoML, facilitate the modeling process and use of predictive models, promoting the democratization of ML. They open up the potential for domain experts to conduct Self-Service ML for the creation and operationalization of predictive models, to improve planning processes and forecasting accuracy. However, the departmental use of Self-Service ML technologies and systems remain under-investigated. For our research, we conducted a case study at an automobile OEM and focused on applications in the field of the operational forecasting and planning of warranty and goodwill costs. Following a case-study approach, our findings suggest that Self-Service ML can be realized but needs a thorough consideration of auditability, interpretability, and support for data provision and model operationalization. We condensed our findings to design requirements and decisions that are supposed to promote practical Self-Service ML implementations and provide a starting point for further research and designing such systems.
Namyslo, Nicole Maria and Baars, Henning, "Harnessing the Business Potential of Self-Service Machine Learning for Forecasting Warranty Costs – Insights from a Case Study in the Automotive Sector" (2022). PACIS 2022 Proceedings. 155.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Paper Number 1319