The problem
Charging hubs vary by site: a kerbside two-bay stop, a mall car park, a highway plaza with dozens of bays. Designing each from scratch is slow and expensive, while one rigid design does not fit sites. Modular architectures promise variety from common parts, but choosing the right module boundaries is hard. This is a research question in product architecture with a strong design and CAD component.
Objectives
- Review modularity and product platform literature and apply it to charging hubs.
- Define functions, interactions and a design structure matrix for a hub.
- Cluster functions into candidate modules and define interfaces.
- Build configurable CAD assemblies and an Alias design language for the modules.
- Generate site variants and evaluate commonality, variety and design effort.
What you need
Tools and toolboxes
- SOLIDWORKS Assembly with Configurations and Design Tables — Configurable modules and hub assemblies generated from a variant table.
- SOLIDWORKS Weldments and Sheet Metal — Canopy, cabinets and frames as configurable structural and sheet parts.
- Autodesk Alias — A shared design language for dispensers, canopy edges and signage, with surfaces exported to SOLIDWORKS.
- SOLIDWORKS Simulation (Static) — Checks of key module interfaces and canopy variants.
- Autodesk VRED or KeyShot — Renders of hub variants in different site contexts.
- Excel or Python — Design structure matrix clustering, commonality indices and variant analysis.
Software
- SOLIDWORKS 2023 or later
- Autodesk Alias 2024 or later
- Excel or Python
Extra parts and data
- Site survey data or photographs for several example sites — Real site sizes and constraints are needed to test whether the architecture covers practical variants.
Key concepts
- Product architecture
- How a product's functions are mapped to physical modules and how those modules interact through interfaces.
- Design structure matrix
- A square matrix showing interactions between elements, such as spatial, energy, information and material. Clustering it suggests module boundaries.
- Commonality index
- A measure of how many parts or modules are shared across variants. Higher commonality usually lowers cost but may reduce fit to each site.
- Interface standardisation
- Fixing the geometry and connections between modules so different modules can be swapped without redesign.
Step-by-step guide
- 1
Review and frame the research
Review literature on modular product architecture, platforms, design structure matrices and mass customisation, and on charging hub design and site planning. Frame questions such as: which module boundaries give the best balance of commonality and site fit for urban hubs, and how much design effort does a configurable CAD system save? Define the variant space from kerbside to highway plaza.
Checkpoint: A literature review, research questions and a defined variant space with example site types.
- 2
Map functions and interactions
List hub functions: grid connection, power conversion, power distribution, dispensing, cable management, user interface, payment, shelter, lighting, energy storage, PV, safety and amenities. List physical elements for each. Build a design structure matrix recording spatial, energy, information and material interactions between elements, with strengths. Validate the matrix with an expert, for example a charging company engineer if available. Score each interaction on a simple scale, for example from minus 2 to plus 2, and record the reason for every strong score so the matrix can be checked.
Checkpoint: A populated design structure matrix with interaction types and an expert review note.
- 3
Cluster into modules
Cluster the matrix with a published algorithm or a clear heuristic in Python or Excel, aiming to keep strong interactions inside modules. Generate two or three candidate architectures, for example a power cabinet module plus dispenser modules, or integrated power-dispenser units. Define interfaces for each: mechanical mounting, power, data, cooling. Compare candidates on interface count and expected commonality. Test how stable the clusters are by changing a few interaction scores and checking whether the module boundaries move.
Checkpoint: Two or three module architectures with defined interfaces and a comparison table.
- 4
Develop a design language
In Alias, design a family look for dispensers, canopy edges, signage and cabinet doors, with shared curves and proportions. Build master surfaces for a dispenser with good continuity, checked with Zebra and curvature evaluation, and create variants for single and dual connectors by changing a few curves. Export surfaces to SOLIDWORKS as STEP. Record the shared curves and proportions as named layers or a separate reference file, so later variants reuse them rather than redrawing.
Checkpoint: A dispenser surface family in Alias with G2 blends and a style guide sheet.
- 5
Build configurable modules
In SOLIDWORKS, model each module with configurations driven by design tables: dispenser with one or two connectors, power cabinet with different power ratings by adding power units, canopy bays repeated by a pattern count, and optional storage or PV modules. Standardise interface geometry as shared reference parts so every module mounts in the same way. Name configurations with a clear code, for example DISP-2C for a two-connector dispenser, so the variant table stays readable as it grows.
Checkpoint: Each module rebuilds in every configuration with interfaces matching the shared reference.
- 6
Assemble site variants
Create a top-level hub assembly with a variant table that selects module configurations and counts for each example site: kerbside two-bay, mall eight-bay and highway plaza with more bays. Place modules on site layouts from real or example site data and check clearances for vehicles, people and service vehicles with simple envelopes. Save each generated variant with Pack and Go as a frozen copy, so later changes to modules do not silently alter past results.
Checkpoint: At least three site variants generated from the variant table, each passing clearance checks.
- 7
Check interfaces and structure
Run static checks on critical interfaces such as dispenser mounting and canopy bay connections, for the most demanding variant, under example wind and impact loads. Confirm that the same interface works across variants. Where one variant fails, decide whether to strengthen the interface for all or allow a variant-specific part, and record the effect on commonality. Run the same checks on the lightest variant as well, because a variant with fewer bays may have less bracing and behave differently.
Checkpoint: Interface checks pass across variants, with any variant-specific parts recorded.
- 8
Measure commonality and effort
Compute commonality indices for each architecture across variants, at part and module level, using a published index or a clearly defined one. Estimate design effort by timing how long it takes to create a new site variant with the configurable system versus modifying a non-modular baseline model, using a small user study with fellow students. Give every participant the same short training and the same site brief, and record errors as well as time.
Checkpoint: Commonality values for each architecture and measured times to create a new variant for both approaches.
- 9
Evaluate fit and trade-offs
Assess how well each variant fits its site using criteria such as used area, number of bays, user walking distance and service access. Plot commonality against site fit for each architecture. Discuss trade-offs, for example integrated units give higher fit for small sites but lower commonality. Gather feedback from practitioners if possible. Weight the fit criteria transparently and repeat the comparison with equal weights to show whether the conclusion depends on the weighting.
Checkpoint: A trade-off chart comparing architectures, with practitioner feedback if available.
- 10
Write and validate
Write the thesis chapter or paper: literature, method, design structure matrix, architectures, CAD system, variants, metrics, user study and limitations. Discuss how results depend on the chosen variant space and matrix. Provide the configurable models and variant table as appendices. Be careful not to claim cost savings without data, and present effort results with their sample size. Include the design structure matrix and clustering code so others can repeat the analysis with their own data.
Checkpoint: A reproducible thesis chapter with appendices and clearly stated limitations.
Testing & validation
- Design structure matrix reviewed by a domain expert.
- Every site variant rebuilds from the variant table without errors.
- Interface checks pass across variants.
- Commonality is measured with a defined index for each architecture.
- Design effort comparison uses a timed user study with stated sample size.
Expected results
A modular charging hub architecture derived from a design structure matrix, a configurable SOLIDWORKS system with an Alias design language, generated site variants and measured commonality, site fit and design effort.
Research angle for your PhD
The contribution is evidence on how module boundary choices for charging hubs trade commonality against site fit, using a design structure matrix method and a working configurable CAD system, with design effort measured in a user study. Publishable work needs an expert-validated matrix, defined metrics and transparent variant data.
Take it further
- Add cost estimates with SOLIDWORKS Costing for each variant.
- Include a battery storage module and study its effect on grid connection size.
- Extend to a hub for electric buses and trucks.
Viva and review questions
How did you choose module boundaries?
By clustering a design structure matrix so strongly interacting elements stay together, then checking interface count and commonality for the candidate architectures.
What is the risk of high commonality?
Modules may be oversized or poorly suited to some sites, adding cost or wasting space. I measured site fit alongside commonality to show the trade-off.
How did you show the configurable system saves effort?
With a small timed user study comparing creating a new site variant from the variant table against modifying a non-modular model, reporting the sample size and spread.
Why use Alias in a structural architecture study?
Modules seen by users need a coherent appearance across variants. Alias gives controlled, high-quality surfaces that can be varied by a few curves and reused across the family.
Learn it at DIYguru
These DIYguru courses and programmes teach what this project needs.
