Farhad Imani

University of Connecticut · United States

Unclaimed · AI-structured profileORCID OpenAlex

Additive manufacturing and robotic inspection researcher at UConn applying agentic AI to process planning and metrology

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Not scored yet: the snapshot gives an h-index of 22 and i10-index of 42, but no verified publication totals or venue-quality breakdown, so Evolve cannot compute a defensible output score. This axis would measure publication volume and productivity against field norms.

Impact—

Not scored yet: no citation totals, per-paper citation distribution, or field-normalized benchmarks were supplied, so impact cannot be derived from the h-index alone. This axis would measure citation-based influence relative to peers in manufacturing automation and inspection.

Focus0/100

Focus scores 0/100, reflecting unusually wide topic coverage: OpenAlex clusters the record across additive manufacturing materials and processes (30 works), 3D printing technologies (27), industrial vision systems and defect detection (24), ultrasound imaging and elastography (18), and photoacoustic imaging (11). Field distribution splits evenly at 0.5/0.5 between manufacturing automation and industrial inspection/robot learning — breadth spanning multiple sensing modalities rather than a single narrow track.

Collaboration41/100

Collaboration scores 41/100 on a moderate footprint: 7 distinct coauthors in the recent three-year window, with a recurring circle of Jingzhan Ge, Ruimin Chen, Azadeh Haghighi, Jiong Tang, Zhiling Chen, Matthew P. Castanier, and David Gorsich. On both reviewed 2026 preprints Imani holds the last author position without corresponding authorship, consistent with a senior supervisory role rather than hands-on execution.

Activity100/100

Activity scores 100/100: OpenAlex logs 47 works in the recent three-year window and an accelerating curve — 10 works in 2022, 19 in 2025, and 18 already in 2026 — the strongest sustained output in the record. The 2026 volume includes two arXiv preprints on agentic robotic manufacturing and inspection, confirming an unusually productive current phase.

Reproducibility20/100

Reproducibility scores 20/100, matching a reproMean of 20 across 2 evaluated papers. Both reviewed 2026 preprints carry a low repro-readiness rating: A-RAM's gains are measured against its own grid's worst candidate rather than external baselines, and FRAME's end-to-end results are small-scale and presented only in figures, limiting independent re-implementation today.

Research summary

Farhad Imani is a researcher at the University of Connecticut whose OpenAlex-indexed record centers on additive manufacturing and 3D printing, industrial vision and defect detection, and ultrasonic imaging. The profile carries an h-index of 22 and an i10-index of 42, with field distribution split evenly (0.5/0.5) between manufacturing automation and industrial inspection/robot learning. Activity is the record's standout axis: 47 works in the recent three-year window and an accelerating curve climbing from 10 indexed works in 2022 to 19 in 2025 and 18 already in 2026. Topic coverage is broad — additive manufacturing materials and processes (30), 3D printing technologies (27), industrial vision systems and defect detection (24), ultrasound imaging and elastography (18), and photoacoustic imaging (11) — a spread that yields a low focus score and suggests a deliberately multi-modal sensing agenda. In 2026, two arXiv preprints entered the Evolve catalog, both with Imani in the last author position: A-RAM, a kinematics-grounded agentic AI planner for robotic additive manufacturing process planning, and FRAME, a measurement-steered vision-language-action inspection framework with deterministic evidence gating. Evolve's reviews credit both with credible engineering cores while flagging weak external baselines and low repro-readiness (reproMean 20 across 2 evaluated papers). Recurring coauthors include Jingzhan Ge, Ruimin Chen, Azadeh Haghighi, Jiong Tang, Zhiling Chen, Matthew P. Castanier, and David Gorsich.

Collaboration pattern

The recent record follows a compact-team pattern: 7 distinct coauthors in the last three years, with a stable recurring circle — Jingzhan Ge, Ruimin Chen, Azadeh Haghighi, Jiong Tang, Zhiling Chen, Matthew P. Castanier, and David Gorsich. On both 2026 catalog preprints Imani takes the last author position without corresponding authorship, fitting a senior-advisor role that frames problems and signs off while collaborators execute. The collaboration score of 41 reflects this small, steady core rather than large distributed networks.

Research keywords

additive manufacturing3D printingrobotic process planningLLM-based agentsindustrial vision systemsdefect detectiondimensional metrologyrobotic inspectionvision-language-action modelsultrasound elastographyphotoacoustic imaginginverse kinematics

Highlights

A-RAM: kinematics-grounded agentic AI planner for robotic additive manufacturing
In this 2026 arXiv preprint (Imani, last author), A-RAM uses an LLM to triage tasks and ground process planning in inverse kinematics, trajectory planning, and joint jerk constraints for robotic AM. Evolve's review finds the kinematics pipeline credible but notes the reported gains are measured against the planner's own grid's worst candidate rather than external baselines; repro-readiness is rated low.
FRAME: measurement-steered VLA manipulation with deterministic evidence gating
This 2026 arXiv preprint (Imani, last author) targets robotic dimensional inspection: vision-language-action manipulation is steered by measurement intent and gated by deterministic metrology checks aligned with laser line profilometry and GD&T/ASME Y14.5. Evolve credits strong ablations while flagging that end-to-end results are small-scale and presented only in figures; repro-readiness is rated low.
Sustained output surge: 10 works in 2022 to 19 in 2025 and 18 in 2026
OpenAlex countsByYear shows the record accelerating from 10 indexed works in 2022 to 19 in 2025 — the annual peak — with 18 already logged in 2026 and 47 works across the recent three-year window. The newest entries pivot toward agentic AI methods for robotic manufacturing and inspection.

Timeline

2011OpenAlex indexes 5 works in 2011, marking the start of sustained output in this record.
2017Indexed output reaches 7 works in 2017, spanning additive manufacturing and ultrasonic imaging topics.
2022A then-record 10 indexed works in 2022 opens a multi-year acceleration phase.
202519 indexed works make 2025 the highest-output year in the record to date.
2026Two arXiv preprints enter the Evolve catalog — the A-RAM agentic planner for robotic AM and FRAME's measurement-steered VLA inspection — both with Imani as last author; 18 works are indexed in total for the year.

Papers in the Evolve catalog

Frequent co-authors

Jingzhan Ge×2 Ruimin Chen×2 Azadeh Haghighi×1 Jiong Tang×1 Zhiling Chen×1 Matthew P. Castanier×1 David Gorsich×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

All bibliometrics are computed from a single OpenAlex-recorded identity; no Scopus, Web of Science, or Google Scholar cross-check was performed.
The output and impact dimensions are unscored because the supplied snapshot lacks verifiable publication totals and citation distributions.
The record contains stray works dated 1988 and 1997 (one each), which may reflect name ambiguity in OpenAlex identity resolution rather than this researcher's activity.
Evolve's paper-level assessment covers 2 arXiv 2026 preprints; the reproducibility score (reproMean 20) derives from these 2 papers and predates any peer review.