Study on Dynamic Value Chain Optimization for the Engineering-to-Order Industry

Abstract

Engineer-to-Order (ETO) environments are characterised by high variability, frequent engineering change, supplier uncertainty, and late-stage field constraints, making traditional linear value-chain and integration-centric execution models increasingly inadequate. While Industry 4.0 initiatives have improved digital connectivity through BIM, ERP, and MES integration, persistent coordination failures and lifecycle blind spots remain unresolved.
This study proposes and evaluates a Dynamic Value Chain Optimisation (DVCO) framework designed to address these limitations through semantic integration, knowledge-graph–based dependency modelling, and AI-enabled reasoning. Using Symbotic as a reference case, the research develops a formal ETO ontology, instantiates a cross-domain knowledge graph, and integrates large language models (LLMs) as semantically constrained decision-support agents. A structured, simulation-driven evaluation compares baseline ETO execution behaviour with a DVCO-enabled environment across engineering, supply chain, manufacturing, installation, and commissioning stages.
The findings demonstrate that DVCO significantly improves lifecycle traceability, change-impact visibility, predictive risk awareness, and cross-functional coordination. Results further show that engineering change and supplier variability are the dominant drivers of non-linear risk propagation, and that DVCO delivers its greatest performance gains in downstream installation and commissioning, where recovery from coordination failure is most difficult. The study concludes that DVCO represents a scalable, technology-agnostic orchestration paradigm capable of transforming ETO execution from a reactive, siloed process into a semantically aligned, model-driven, and AI-supported value chain.