TEC4CPC – Towards a Trustworthy and Efficient Companion for Car Part Catalogs

Patrick Lang

B.Sc. at N4

N4, a leading provider of procurement platforms in the automotive sector, is facing the challenge of making its catalogs for car parts (N4Parts) more user-friendly. These catalogs are used by customers both to purchase parts and to obtain information, such as installation instructions and maintenance intervals. The use of such browser-based applications often requires specific automotive knowledge and different digital skills from users. Since all product information is provided by master data suppliers in a predefined format, it is difficult for N4 to ensure the accuracy of the data or to customize the presentation of the core data. A promising approach to solving this problem is the use of a chatbot. Such a bot could allow users to easily query information without having to navigate through the system. However, the implementation of a chatbot also entails risks: the response quality can vary and for some users the use of a chatbot can be less efficient than conventional navigation. The aim of the TEC4CPC research project is to develop and evaluate a chatbot that will make interaction with N4Parts more efficient and of higher quality. The chatbot uses a Large Language Model (LLM) in combination with Retrieval Augmented Generation (RAG) to provide accurate and up-to-date answers. Such a system could particularly benefit new employees who need to access the information they need quickly and efficiently. In addition, digital companions could improve the user experience by enabling standardized processes and less manual work. In the long term, N4 aims to develop adaptive user interfaces that suggest the most efficient method of interaction based on the user’s behavior and previous experience. This project lays the foundation for future collaborations between N4 and the German Research Center for Artificial Intelligence (DFKI) to further optimize the entire N4 AUTOMOTIVE SUITE.

Keywords: Automotive Procurement Systems, E-commerce Solutions for Automotive Industry, Car Parts Catalogs, Digital Media Usability, User Experience (UX) in Automotive Applications, Human-Computer Interaction (HCI), Information Retrieval Systems, Data Integration and Master Data Management, Natural Language Processing (NLP) in Automotive, Chatbot Implementation in E-commerce, User Skill Level Adaptation, Knowledge Representation in Automotive Systems, Usability Challenges in Browser-based Applications, Data Accuracy and Liability in E-commerce, Intelligent Assistants in Procurement Platforms, User-Centric Design for Digital Interfaces, Efficiency of Digital Tools in Automotive Procurement, Comparative Analysis of Chatbot and Traditional Navigation, Automation and User Assistance Systems, Data-driven Decision Making in Automotive E-commerce

Scientific area: Artificial Intelligence, Knowledge representation and reasoning

Bio: Patrick Lang is an accomplished IT expert and department head in the field of e-commerce solutions for the automotive industry. His career began after completing his degree in computer science, followed by various positions in IT development and project management.

Visiting period: June-August ’24 at DFKI