Decentralized additive manufacturing can help manufacturers produce components closer to end users and respond more flexibly to demand. But distributing production across facilities creates a technical challenge: a part designed for one machine, material and process setup may not deliver the same result elsewhere. Repeatability depends on understanding how component geometry interacts with each facilityâs equipment and manufacturing constraints.
A model-based design method offers a structured way to manage those differences. By connecting functional requirements, design decisions, manufacturing data and testing across the product lifecycle, teams can assess alternatives for multiple sites rather than treating each production location as an isolated operation.
Why decentralized additive manufacturing needs connected data
Additive manufacturing (AM) can reduce supply-chain stages and support on-demand production. A distributed network may also locate manufacturing closer to customers. However, the component outcome can vary with the machine, process parameters, material and geometry. Without a common method for capturing and sharing relevant information, a design that works at one facility may require additional post-processing or prove unsuitable at another.
Data silos make the problem harder. Simulation, design and production information may be collected at separate stages without being available to the teams making decisions elsewhere in the lifecycle. This limits traceability and quality assurance, and makes it difficult to use production experience to guide subsequent design work. In a multi-partner network, missing or poorly documented data can also lead to repeated testing.
The practical need is not simply to store more data. Teams need to make relevant lifecycle information accessible during design, where it can inform geometry, process choices and trade-offs such as cost, weight and manufacturing time.
A model-based design method for decentralized additive manufacturing
The proposed method organizes development into five connected steps. It uses a function model to document requirements and constraints, then iterates between design, analysis and targeted testing before releasing the component for production.
- Build a function model. Identify what the component must do, its design specifications and the constraints that affect them. These include functional requirements as well as manufacturing limits, such as minimum wall thickness.
- Redesign for AM and analyse. Develop an additive design and assess uncertainties related to geometry, manufacturing capability and quality. Because constraints vary between machines and processes, the analysis must consider each intended production facility.
- Manufacture and analyse test artefacts. Where information is missing, create targeted test pieces to investigate specific uncertainties. Their production conditions should match the relevant component conditions, including machine parameters, material and build-plate position.
- Update the model and iterate. Use test results to refine the documented constraints and compare design alternatives. Repeat the design-analysis-testing cycle until the relevant uncertainties are addressed.
- Manufacture the selected design. Once the design has been assessed against the identified uncertainties and the team has selected an alternative, the component can be produced at the intended facilities.
The function model acts as a shared reference for requirements and manufacturing information. In the method, the Enhanced Function-Means approach is used because it can represent both functions and constraints; other modelling approaches with similar capabilities may also be suitable.
Use targeted testing to resolve design uncertainty
General design guidance and machine documentation do not always provide the information needed for a particular component. A test artefact can fill that gap by exploring a specific relationship, such as how surface inclination affects roughness or post-processing needs. The results can then inform design parameters and help teams evaluate their effects on cost or manufacturing time.
This approach makes testing more purposeful: rather than relying only on broad campaigns, teams identify uncertainties during design and build tests to address them. Test results should be documented in the model so they can inform later analysis and be communicated across participating design and manufacturing teams.
Trade-off analysis helps make the choices visible. For example, a design optimized for the least restrictive machine might require extra post-processing at a more restrictive facility. Designing to the most restrictive machine may ease production there but constrain alternatives. A compromise design may balance the needs of all sites. The preferred option depends on the projectâs priorities, which may include manufacturing cost, production time or weight.
What the steering-knuckle case demonstrates
The method was illustrated by redesigning a car steering knuckle for production at three facilities in Sweden and Italy. The team used a function model to record component requirements and manufacturing constraints, then developed an AM design using topology optimization and design iterations intended to reduce support structures.
Two design considerations required further investigation: surface inclination, which affects surface roughness, and wall thickness, which can matter where rough surfaces or material removal are involved. The available information did not establish the relationship between inclination and roughness for the machines in the project. The team therefore designed test artefacts to gather missing data and refine the design analysis.
Design alternatives and printing orientations were compared using measures including weight, estimated manufacturing cost and manufacturing time. The case illustrates why decisions need to account for the complete operation, including printing and post-processing, and why the best design depends on the objectives of a particular application. It does not establish that one alternative will be optimal for every component or production context.
From lifecycle data to repeatable production
The methodâs central contribution is a process for bringing manufacturing knowledge into design decisions across multiple facilities. A documented model can help teams identify uncertainty, target testing and share results, supporting more informed choices about component geometry and production trade-offs.
It is intended to complement existing data-management tools, not replace them. The case study also exposed practical limitations: analysis was performed manually across several software tools, and adding manufacturers increased the effort involved. A more integrated platform and digital thread could make lifecycle information easier to retrieve and support more automated design analysis. The method was demonstrated in one case, so its broader applicability requires further validation.
For manufacturers and integrators planning distributed AM, the practical implication is clear: treat design, process capability and data management as connected concerns. Model requirements across facilities, test the uncertainties that matter, and record the results where they can inform future decisions. That discipline can provide a stronger basis for repeatable production without assuming that every machine or site will behave identically.
Access Official Technical PDF