29.06.2026 | Contributions at FAIM 2026
The Chair of Application Systems and E-Business is represented with two contributions at the International Conference on Flexible Automation and Intelligent Manufacturing (FAIM 2026):
Tamino Marahrens presented the papers:
A Domain-Specific Data Model for Asset-Centric Digital Twins in Intralogistics
Abstract: Intralogistics systems rely on heterogeneous physical assets such as storage systems, conveying technologies, and transport vehicles whose states and interactions evolve dynamically during operation. Although digital twins (DTs) are increasingly applied to support monitoring, simulation, and decision-making in these environments, many existing implementations remain tightly coupled to specific applications or processes. As a result, the underlying data structures required for interoperable and reusable asset-level DTs are often implicitly embedded in system solutions rather than explicitly modeled. This paper addresses this gap by proposing a domain-specific conceptual data model for asset-centric DTs in intralogistics. The approach defines the asset twin as a persistent digital representation of a specific physical asset and structures asset-related information into three complementary elements: asset identification, asset description, and associated information objects. In addition, the model differentiates between static, dynamic, and lifecycle-dependent data to capture the temporal evolution of asset information across operational and lifecycle contexts. By separating stable descriptive attributes from operational states and event-based lifecycle records, the proposed structure enables consistent referencing, reduces process coupling, and supports interoperable use of asset information across heterogeneous systems. The resulting model contributes to a more structured and reusable asset-level conceptualization of DTs in intralogistics.
How Good is my Scheduling Algorithm? A Benchmark to Compare Production Scheduling Algorithms for Real-World Production Environments
Abstract: Production scheduling in modern manufacturing faces escalating complexity due to mass customization, frequent disruptions, and stringent lead-time requirements, rendering classical benchmarks inadequate. While a large body of scheduling algorithms has been proposed to cope with this complexity, their evaluation is hampered by the lack of a shared, realistic benchmark: established instance sets abstract from real-world shop‑floor constraints, and studies that do account for practical constraints typically rely on self‑created scenarios that are not comparable across approaches. As a result, it remains difficult to judge the relative solution quality of scheduling algorithms in industrially relevant settings. This paper introduces a realistic, multi‐instance benchmark to evaluate and compare scheduling algorithms under real-world shop‐floor constraints. Guided by a problem-centered design-science methodology, we derive high-level requirements—sufficient complexity, differentiation capability, objective-function variants, and resistance to overfitting—and implement ten real-world-relevant features, including sequence-dependent changeovers, limited buffers, personnel constraints, transportation resources, and job release dates. Performance is assessed via average relative improvement over a “random” priority-rule baseline, which mitigates randomness and discourages overfitting. We demonstrate and validate the benchmark through the application of FIFO, due-date, changeover heuristics, and a reinforcement-learning algorithm, showing clear differentiation among methods. We provide the benchmark and all necessary data for subsequent use in the supplements.
