Transform Fragmented Inventory Data into Coordinated Action

PRISM (Process Intelligence for Replenishment, Inventory, and Supply Management) applies state-of-the-art object-centric process mining to give your operations live awareness and explainable decision support.

Explore Partnership Opportunities

The Challenge: Fragmented Operations

Manufacturing and logistics organizations do not suffer from a lack of data. They suffer from fragmentation. Information is scattered across ERP, WMS, MES, and quality systems, while operational consequences appear only later as shortages, excess stock, or delayed replenishment.

The PRISM Solution

PRISM aims to help companies manage inventory as a process rather than a static stock figure. We connect heterogeneous operational sources and translate them into practical, implementation-oriented outcomes.

Earlier shortage detection & risk scoring
Better replenishment timing & lower overstock
Faster exception handling & cross-company coordination

PRISM Overall Architecture Outline showing data flowing from fragmented operational systems to coordinated actionable outcomes

Explainable & Feasible Interventions

PRISM doesn't just monitor data; it drives action. Our open-source architecture consists of mapping (MAP), live sensing (PULSE), collaboration (SHARE), and decision support (ACT).

PRISM ACT is our dedicated decision and intervention layer. It converts live process signals into ranked, explainable intervention options. Planners and warehouse managers are supported in choosing the right response—such as reallocation, reprioritization, or substitution—when inventory risks emerge, managing trade-offs consciously before execution.

PRISM ACT workflow showing incoming signals converting into a ranked and explainable intervention set

The Team Behind PRISM

PADS and Fraunhofer FIT combine process mining research, open-source software, and industrial transfer to make process intelligence usable in real operations.

The PRISM Project Caretakers: Prof. Wil van der Aalst, Dina Kretzschmann, Dr. Alessandro Berti, Humam Kourani

Are you interested in a (substantially) funded research collaboration? Please contact us.

We are actively looking for industrial partners with real-world inventory, warehouse, or material flow challenges to validate and scale our open-source capabilities.

Project Caretakers:

Dina Kretzschmann
dina.kretzschmann@pads.rwth-aachen.de

Dr. Alessandro Berti
a.berti@pads.rwth-aachen.de

Humam Kourani
humam.kourani@fit.fraunhofer.de

Selected Research Output

Alessandro Berti, Dina Kretzschmann, et al. "Interpretable Execution State Abstraction from Object-Centric Event Logs for Process-Aware Decision Support: Linking Execution States to Stock and Policy Regimes" Accepted at HybridAIMS 2026 (CAiSE 2026) [PDF]
Alessandro Berti, Dina Kretzschmann, et al. "From Object-Centric Event Data to Causal Inventory Insights: A Structural Equation Model of Understock and Overstock." Accepted at AILS 2026 (BIS 2026) [PDF]
Dina Kretzschmann, Alessandro Berti, et al. "State-Aware Object-Centric Process Mining: Enhancing OCEL 2.0 with Explicit State Transitions." EDOC 2025. [PDF]
Alessandro Berti, Dina Kretzschmann, et al. "Segmentation for Optimizing Long-Lifecycle Processes in Object-Centric Process Mining." Accepted at the AKTB 2025 workshop (BIS 2025). [PDF]
Dina Kretzschmann, Alessandro Berti, et al. "A Data-Driven Framework for Retail Inventory Optimization: Integrating Object-Centric Process Mining and Mathematical Models." Accepted at BIS 2025. [PDF]
Dina Kretzschmann, et al. "Overstock Problems in a Purchase-to-Pay Process: An Object-Centric Process Mining Case Study." International Conference on Advanced Information Systems Engineering. Cham: Springer Nature Switzerland, 2024. [PDF]
Alessandro Berti, et al. "Analyzing Inter-Connected Processes: Using Object-Centric Process Mining to Analyze Procurement Processes." International Journal of Data Science and Analytics (2023). [PDF]
Alessandro Berti, Wil M.P. van der Aalst. "Retrieval-Augmented In-Context Foundation Model for Predictive Process Monitoring" Pre-Print. [PDF]
Anton Antonov, Humam Kourani, Alessandro Berti, et al. "PMAx: An Agentic Framework for AI-Driven Process Mining" Accepted at EMMSAD 2026 (CAiSE 2026). [PDF]
Alessandro Berti, Wil M.P. van der Aalst. "An In-Context Foundation Model for Predictive Process Monitoring on Event Logs" Accepted at IEEE Access. [PDF]
Alessandro Berti, Xiaoting Wang, Humam Kourani, Wil M.P. van der Aalst. "Specializing Large Language Models for Process Modeling via Reinforcement Learning with Verifiable and Universal Rewards" Process Science. [PDF]
Anton Antonov, Humam Kourani, Alessandro Berti. "Beyond Control Flow: Integrating Organizational Perspectives into Generative Process Modeling" Pre-Print. [PDF]
Alessandro Berti, Humam Kourani "Diagnosing LLM Hallucinations in Process Mining Tasks: a Taxonomy and a Benchmark" Pre-Print. [PDF]
Humam Kourani, Anton Antonov, Alessandro Berti, Wil M.P. van der Aalst. "Knowledge-Driven Hallucination in Large Language Models: An Empirical Study on Process Modeling" Accepted at GenAI4PM 2025, 2025. [PDF]
Alessandro Berti, Humam Kourani, Gyunam Park, Wil M.P. van der Aalst. "Configuring Large Reasoning Models using Process Mining: A Benchmark and a Case Study" NLP4BPM Workshop, BPM 2025. [PDF]