SAP
SAP is the leading enterprise application and business AI company. SAP stands at the intersection of business and technology, where innovations are designed to directly address real business challenges and produce real-world impacts. SAP’s integrated portfolio unites the elements of modern organizations — from workforce and financials to customers and supply chains — into a unified ecosystem that drives progress. For over 50 years, organizations have trusted SAP to bring out their best by uniting business-critical operations spanning finance, procurement, HR, supply chain, and customer experience. For more information, visit www.sap.com.
SAP and AI
SAP focuses on Business AI by embedding intelligent capabilities directly into enterprise applications and processes from the start. These solutions meet SAP’s high standards for product quality, ethics, privacy, and security.
Using advanced machine learning and deep learning technologies, SAP enhances its cloud solutions to deliver practical business value. For example, this technology/AI helps automatically match invoices to payments in CloudERP SAP Cash applications, streamline invoice processing in SAP Concur within Spend Management.
Predictive Community x IUC Applied Research Talks
The Applied Research Talks offer a series of talks where experts from SAP and TUM provide in-depth insights into our ongoing collaborative applied research projects and further interesting research topics. These sessions serve as a bridge between SAP and TUM, fostering collaboration and knowledge exchange in an interactive setting. Each talk delivers a spark of innovation—introducing fresh perspectives on emerging technologies and research-driven advancements that shape the future of enterprise solutions. Details here
"Synthetic Business Data Generation" with Prof. Stephan Günnemann
Project Title: "New Approaches for Synthetic Business Data Generation"
Conducting research does not only require well-defined problem statements but particularly in AI also relevant data. However, SAP as a leading ERP provider, cannot just share confidential customer data with third parties and researchers. While public dataset may be used it is often not representative of real business data. To this end, this project aims to synthetically generate business data driven by SAP data models and metadata, allowing for both scientific reproducibility and business relevance.
"New Approaches to Tour Planning and Vehicle Routing" with Prof. Stefan Minner
Project Title: “New Approaches to Tour Planning and Vehicle Routing”
This is a joint research project regarding the use of data-driven approaches and decomposition algorithms in the context of the Vehicle Routing Problem (VRP) to enable the solution of larger and more complex real-world problems. In order to advance the state of knowledge in AI and advanced optimization, several more specific questions need to be addressed. For example, how are the clusters rated when aggregating and breaking down into sub-clusters, and how to improve the user acceptance of data-driven and AI-based approaches.
"Probabilistic Planning for Resilient Supply Chains" with Prof. Martin Grunow
Project Title: "New Approaches for Stochastic / Range-Based Planning and Simulation for Resilient Supply Chains"
This project focuses on supply chain planning and design, with an emphasis on advanced approaches for stochastic and range-based planning and simulation to support resilient supply chains. Key questions include increasing robustness through stochastic methods and gaining insights into probabilistic planning strategies for resilient supply chains.
"AI in Production Planning" with Prof. Martin Grunow
Project Title: "AI for Production Planning and Detailed Scheduling"
This research investigates production planning and detailed scheduling in stochastic environments. The primary objectives are to utilize machine learning (ML) for identifying suitable approaches to production scheduling under uncertainty, develop frameworks for integrating online decision-making and planning, benchmark novel ML approaches against classic methodologies, match methodologies to specific production settings, and integrate online decision-making into detailed scheduling processes.
"Building Semantic Models for the Process Mining Pipeline" with Prof. Stefanie Rinderle-Ma
Project Title: "Building Semantic Models for the Process Mining Pipeline"
This project explores process mining as a method of querying process data from event logs to improve business processes. It aims to reduce the human effort required in preparing analyses and to develop semantic models for process mining recommender systems. The central question is whether expert knowledge used for analysis generation can be produced, exploited, and scaled effectively. A key challenge lies in integrating human and machine-driven efforts to make process mining analyses reusable across organizations.
"Enterprise Data Science and AI" with Prof. Florian Matthes
Project Title: "Applications of Text Generation Through Semi-Supervised Learning and Other State-of-the-Art Machine Learning Approaches"
This project explores the application of semi-supervised learning frameworks in text generation using deep learning models. Research questions include automated labeling of data generated from chatbot interactions, integrating user feedback to enhance chatbot learning experiences, and improving chatbot responses when training data is limited. The project evaluates state-of-the-art methods in semi-supervised learning and natural language processing, including the use of LLMs and optimizing GenAI architectures, for example, through Retrieval Augmented Generation (RAG), for diverse applications and use cases.