The project is a direct collaboration between the Chair of Machine Tools and Control Systems at the Technical University of Kaiserslautern-Landau and the Chair of Artificial Intelligence and Intelligent Information Systems at the University of Trier. Together, the two chairs focus on the following objectives:

• Enhancing Internal Networking and Data Utilization:
  Developing a unified platform to integrate data from all production stages, optimizing information flow and performance in the product development process. Dynamic decision support is employed to sustainably increase value creation.
• Incorporating Sustainability into Design and Production:
  Developing formal methodologies to include factors such as time, material, energy, and circular economy principles in product development and subsequent production phases. The goal is to minimize waste, use resources more efficiently, and promote sustainable production.
• Identifying Optimization Potential in the Design Process:
  Analyzing and assessing complexities in design decision-making using digital models such as Digital Twins. This approach allows for dynamic adaptation of production processes based on standardized interfaces and contextual parameters.
• Developing a Hybrid AI-Based Decision Support System:
  Designing an interactive system that provides product designers with explainable AI recommendations and simulations. The system incorporates human expertise (Human-in-the-Loop) to avoid errors and optimize the design process.
• Practical Demonstration on a Real Production Line:
  Extending an existing truck assembly line with decision-support functionalities based on Federated Learning (FL) and Case-Based Reasoning (CBR). This aims to create a practical application scenario for adaptive quality control and sustainable production.

The product development process is a complex, iterative workflow spanning from concept to final prototype. Designers face numerous decisions impacting complexity, manufacturability, and resource requirements. Traditionally, the process relies heavily on digital tools like databases and on the experience gained from prior projects.

A key goal is to employ Decision Support Systems (DSS) to optimize and streamline the development process. Artificial Intelligence (AI) serves as an enabling technology, offering explainable decision-making aids. Although initial efforts have been made to apply AI in product development, its use remains in its early stages.

• WP 1: Data Science – Product Design and Manufacturing Processes:
  Investigating machine-readable information models and process descriptions for automated traceability and improvement of product design processes.
• WP 2: Specification of the Interactive Decision Support System:
  Specifying the hybrid Interactive Decision Support System, the core of the XDP-Opt project.
• WP 3: Federated Learning Methods for Quality Control:
  Using FL methods, a specialized form of machine learning, to gather quality data and develop a global model, ensuring data confidentiality through decentralized architecture.
• WP 4: Case-Based System for Design Decisions:
  Developing a CBR system to capture and reuse past design decisions and their impact on production processes and quality.
• WP 5: AI-Based Exploration of the Product Design Space:
  Exploring components that dynamically respond to their environment and provide designers with options and recommendations based on the design space.
• WP 6: Provision and Distribution of Data:
  Distributing existing data and generating additional data based on use cases, with open access for SPP partners and the public.
• WP 7: Development, Demonstration, and Evaluation of the Interactive Decision Support System:
  Demonstrating and experimentally evaluating the complete approach using the truck assembly line use case at SmartFactoryKL.