Generative design and topology optimization in 2026: overview
Generative design and topology optimization use computational methods to create or refine designs against constraints such as loads, material use and manufacturing limits. Rather than starting with a fixed shape, engineers define a design space and objectives; algorithms search for structures that meet them. The approach can support lighter, more efficient components, but its value depends on whether designs can be analyzed, manufactured and validated. Current work increasingly connects optimization with fabrication constraints, material behavior and performance across multiple physical domains.
Recent research is extending both the algorithms and their engineering use. In September 2026, a Santa Clara University study used a large language model to adapt continuation parameters during SIMP optimization, while Nanjing University of Science and Technology researchers applied random-forest regression to thin-walled structures. In August 2026, University of Michigan-Dearborn researchers published work on generative liquid-cooling channels for advanced chip packaging. These efforts point toward faster, more application-aware search, while reliability, manufacturability and computational cost remain open challenges.
- 1AI enters the solver: Recent work uses language models and learned surrogates to guide or accelerate optimization, while keeping physical constraints central to the design process.
- 2Thermal demand rises: Thermal management and electronics is a growing application area, with recent work on liquid-cooling channels for advanced packaging and optimized heat-dissipation structures.
- 3Makeability matters: Research increasingly integrates manufacturing limits, material orientation and reliability into optimization rather than treating fabrication as a separate downstream step.
Top organizations working on generative design and topology optimization
Ranked by Linknovate activity score · last 24 monthsThe activity score adds up the relevance of every patent, publication, grant and news item linking an organization to generative design and topology optimization. Showing 7 of 1,358 organizations.
| # | Organization | Approach | Signals (24 mo) | Momentum | Why it matters |
|---|---|---|---|---|---|
| 1 | 
Dalian University of Technology
Research · China |
Density-based methods | PUB 196 | +8% | Recently, researchers developed a Voronoi-based framework for optimizing manufacturable porous infill and a method for jointly optimizing structural topology and material orientation. |
| 2 | 
Huazhong University of Science and Technology
Research · China |
Level-set methods | PUB 132 · NEWS 13 | -1% | Recently, researchers developed level-set approaches for macro–micro cellular structures and auxetic metamaterials, including experimental analysis of additively manufactured designs. |
| 3 | 
Northwestern Polytechnical University
Research · China |
Density-based methods | PUB 91 | +46% | Recently, researchers developed robust topology optimization methods for multi-material heat exchangers and structures made from materials with different tensile and compressive properties. |
| 4 | 
Xi'an Jiaotong University
Research · China |
Density-based methods | PUB 81 | +8% | Recently, researchers applied topology optimization to wireless power transfer magnetic cores, acoustic resonators, and turbulent heat-transfer microchannels. |
| 5 | 
Beihang University
Research · China |
Evolutionary methods | PUB 70 | -11% | Recently, researchers developed reliability-based multiscale topology optimization for aircraft lattice structures, incorporating overhang control to address manufacturability. |
| 6 | 
Beijing Institute of Technology
Research · China |
Generative and AI-assisted search | PUB 72 | +12% | Recently, researchers proposed a physics-guided diffusion model with differentiable connectivity constraints for topology optimization, targeting faster design exploration while retaining physical guidance. |
| 7 | 
Harbin Institute of Technology
Research · China |
Evolutionary methods | PAT 1 · PUB 72 | +43% | Recently, researchers applied collaborative gradient optimization to glass-sponge-inspired lattice structures and developed inverse design for snapping structures with specified mechanical responses. |
| 8 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 9 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
| 10 | ··Hidden organizationCompany | Approach | PAT 00 · PUB 00 GRANTS 0 · NEWS 00 | +00% | Available to registered users. |
Startups to watch and leading research centers
Startups to watch 16 total
DiabatixBelgiumDiabatix develops generative-design software for heat exchangers and heat sinks, using numerical optimization to tailor cooling geometries to thermal and flow requirements.
Renaissance FusionFranceRenaissance Fusion develops stellarator technology; topology optimization has been applied to the structural design of a high-field stellarator magnet.
Solenic MedicalUnited StatesSolenic Medical is developing magnetic-field treatment for prosthetic joint infections; a cited patent describes topology optimization of a coil winding pattern for a medical transducer.- ··Hidden startupCountryAvailable to registered users, with funding and founding year.
- ··Hidden startupCountryAvailable to registered users, with funding and founding year.
Leading research institutions 1,166 total
- Dalian University of TechnologyChinaRecently, researchers developed a Voronoi-based framework for optimizing manufacturable porous infill and a method for jointly optimizing structural topology and material orientation.
- Huazhong University of Science and TechnologyChinaRecently, researchers developed level-set approaches for macro–micro cellular structures and auxetic metamaterials, including experimental analysis of additively manufactured designs.
- Northwestern Polytechnical UniversityChinaRecently, researchers developed robust topology optimization methods for multi-material heat exchangers and structures made from materials with different tensile and compressive properties.
- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
- ··Hidden research institutionCountryAvailable to registered users, with their key researchers.
Where generative design and topology optimization innovation happens
Share of documents with at least one organization from each country.
Open the interactive map with all 1,358 organizationsMain technical approaches
Last 24 months| Approach | What it is | Documents | Leading organizations |
|---|---|---|---|
| Density-based methods | Density-based methods vary material presence across a design grid and work well with established solvers, but can require filtering and post-processing to produce clear, manufacturable boundaries. | 558 |
Huazhong University of Science and Technology, Dalian University of Technology, Northwestern Polytechnical University |
| Level-set methods | Level-set methods represent boundaries as evolving surfaces, yielding clear interfaces but making topology changes and complex multi-material designs more difficult. | 256 |
Huazhong University of Science and Technology, University of Tokyo, Dalian University of Technology |
| Evolutionary methods | Evolutionary methods iteratively add or remove material based on performance, offering an intuitive search process but often requiring many structural analyses. | 126 |
RMIT , Széchenyi István University, Tongji University |
| Ground-structure methods | Ground-structure methods select members from a predefined network, making structural layouts explicit but limiting results to the network’s resolution and geometry. | 36 |
Tongji University, Chongqing University, Dalian University of Technology |
| Generative and AI-assisted search | Generative and AI-assisted search learns or proposes candidate designs quickly, but results must be checked for physical validity, constraint compliance and manufacturability. | 1,728 |
Dalian University of Technology, Generative Design, Huazhong University of Science and Technology |
Emerging trends
Last 12 months vs previous 36 monthsTerms whose frequency in the last 12 months grew the most compared with the previous 36 months.
Funding and investment
Recent funding rounds
Since 2020, the field has received $69,450,809 across 115 grants. The available funding data does not establish that every grant is dedicated solely to generative design or topology optimization.
Patent landscape
Top patent holders · last 24 monthsRecent developments
Latest news, duplicates removed- Aug 2026Researchers from University of Michigan-Dearborn published a technical paper titled “Generative Design of Liquid-Cooling Channels for Thermal Management of 2.5D and 3D Integrated Advanced Packaging.” Abstract Excerpt…Semiconductor Engineering
- Aug 2026The Federal Energy Regulatory Commission on Wednesday approved the Southwest Power Pool’s proposal for using “topology optimization” to reduce grid congestion and its related costs. Traditionally, grid operators respond…Utility Dive - Latest News
- Jul 2026The drive to reduce weight while maintaining performance continues to shape engineering decisions across aerospace, motorsport and other high-performance industries Generative design and topology optimisation have…Engineer Live
- Jun 2026In 2022, global production of construction materials accounted for more than 7 percent of total carbon emissions. But how many of those materials were truly necessary to build houses, buildings, and bridges? A technique…MIT News
- Month 20XXHidden developmentAvailable to registered users with alerts enabled.
Frequently asked questions about generative design and topology optimization
What companies are working on generative design and topology optimization?
Autodesk, Arteris Inc., JFE Steel and Siemens AG appear among the leading patent holders in the field. Diabatix develops generative-design software for heat exchangers and heat sinks. The available organization data also identifies research groups and software-related activity, but patent counts alone do not show which firms offer commercial products or services.
Which generative design and topology optimization startups are the most promising in 2026?
Diabatix is a named startup focused on generative-design software for heat exchangers and heat sinks, using numerical optimization to meet thermal and flow requirements. Renaissance Fusion and Solenic Medical also appear in the startup data, though topology optimization is an application within their broader work. The supplied information does not provide comparable commercial or investment metrics to rank these companies.
Which universities and research centers lead generative design and topology optimization research?
Dalian University of Technology leads the listed recent research organizations with 196 documents in the window, followed by Huazhong University of Science and Technology with 145 and Northwestern Polytechnical University with 91. Their work includes porous infill, cellular structures, auxetic metamaterials, multi-material heat exchangers and robust optimization under different material behaviors.
How much funding has generative design and topology optimization received?
The available data records $69,450,809 across 115 grants since 2020. These figures describe grants associated with the field’s data, not a verified total of funding exclusively dedicated to generative design or topology optimization. Separate funding-round entries include activity across organizations and sectors, so they should not be treated as a clean field-specific investment total.
Who holds the most generative design and topology optimization patents?
Autodesk leads the listed patent holders with 13 patents in the dataset. Arteris Inc. and JFE Steel follow with 7 each. Siemens AG and Honda Corporation each have 3. These counts identify organizations associated with patents in the field’s records; they do not measure the quality, commercial value or current status of those patents.
Is generative design and topology optimization patent activity growing?
Patent activity is down: 104 patents appear in the 24-month window, and patent growth is -34.9% for the last 12 months versus the previous 12 months. Publication activity moved in the opposite direction, with 2,508 publications last year and growth of 3.3%. This contrast suggests continuing research activity despite a recent decline in patent output.
What are the latest developments in generative design and topology optimization?
In September 2026, Santa Clara University researchers presented an LLM-based controller that adapts continuation parameters during SIMP optimization. Also in September 2026, Nanjing University of Science and Technology researchers applied random-forest regression to thin-walled structure design. In August 2026, University of Michigan-Dearborn researchers published generative design work on liquid-cooling channels for advanced chip packaging.
What are the main approaches in generative design and topology optimization and how do they differ?
Density-based, level-set, evolutionary, ground-structure, and generative and AI-assisted search methods are the main approaches. Density-based methods vary material across a grid; level-set methods evolve boundaries; evolutionary methods add or remove material; and ground-structure methods select members from a predefined network. AI-assisted methods propose or learn designs, but still require physical and manufacturing checks.
What are the applications of generative design and topology optimization?
Thermal management and electronics is the largest listed application area, with 477 documents, followed by aerospace and automotive with 173 and additive manufacturing with 143. Other applications include biomedical devices, with 112 documents, and civil infrastructure and architecture, with 99. Recent examples include cooling-channel design, lightweight structures and patient-specific or compliant devices.