From Capability Models to Automated Planning: An AAS-Native Approach for Automatic PDDL Generation

Researchers have developed an automated method to generate PDDL planning problems directly from Asset Administration Shell (AAS) capability models using Industry 4.0 standards, eliminating the need for specialized planning expertise. This approach enables production engineers to design and verify manufacturing system layouts without requiring knowledge of formal planning languages, significantly reducing barriers to adopting automated planning in industrial settings.

This research addresses a critical bottleneck in Industry 4.0 adoption: the expertise gap between production engineers and automated planning specialists. Manufacturing facilities require verification that their layouts can execute all necessary production sequences, yet PDDL formulation demands specialized knowledge most engineers lack. The work bridges this gap by leveraging existing standardization frameworks already embedded in industrial digital twins, making planning accessible to domain experts without formal computer science training.

The approach builds on the Asset Administration Shell framework, which has gained traction as the standardized digital twin representation for Industry 4.0 systems. By extracting planning elements from capability models structured according to four established standards—VDI 3682, IEC 61360-1, IDTA 02011, and IDTA 02016—the method creates complete PDDL problems automatically. This represents a paradigm shift from forcing engineers to learn planning syntax toward letting them work in their native language of industrial capability descriptions.

The practical impact extends across manufacturing design and operations. Engineers can now systematically explore design trade-offs by modifying AAS models and regenerating planning domains, accelerating the iterative design process. The validation on a laboratory production system demonstrates the approach handles real-world complexity across multi-AAS architectures. This democratization of automated planning could substantially increase adoption rates in manufacturing automation, reducing design errors and optimization cycles.

Looking forward, the success of this automatic generation approach may inspire similar abstractions in other industrial domains. The broader implication suggests that as digital twins become standard infrastructure, layers of automation will increasingly emerge to translate domain models into analytical frameworks, further reducing manual specialized work in industrial engineering.

• →Automated PDDL generation from AAS capability models eliminates the need for planning expertise in production engineering
• →The method leverages four established Industry 4.0 standards to extract sufficient information for complete planning problem formulation
• →Engineers can now explore manufacturing layout trade-offs directly through model modifications without learning formal planning syntax
• →The approach handles distributed multi-AAS architectures, scaling to realistic industrial system complexity
• →This pattern of automatic code/model generation from domain standards may accelerate digital twin utility across industrial applications