Industrial ML researcher at Hitachi — reinforcement learning for predictive maintenance, LLM agents for supply chains.
2
papers here
2
reviewed by Evolve
2
first-author
0
corresponding
—
works · OpenAlex
12
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 yet scored: the current data does not include a complete verified publication record for this researcher, so raw output volume cannot be computed. This axis would measure total works, yearly output distribution, and output tier against field peers once a full record is ingested.
Impact—
Not yet scored: citation-level data is insufficient in the current snapshot, so influence cannot be quantified. This axis would measure citations received, citing works, and benchmark standing against peers in the same fields; no estimate is offered here.
Focus0/100
The radar focus score is 0, reflecting an evenly split record: Evolve's field distribution assigns 0.5 to reinforcement learning for predictive maintenance (PHM) and 0.5 to LLM-based multi-agent systems for enterprise supply chain analytics. OpenAlex topic counts reinforce the spread — reinforcement learning in robotics (11), adversarial robustness (6), industrial vision and defect detection (6), neural network applications (5), and machine fault diagnosis (5). This reads less as scattered work than a deliberate two-track industrial AI agenda.
Collaboration36/100
Collaboration scores 36 on the radar. The recent three-year window shows 13 works with only 5 distinct coauthors, and the recurring partner list is short: Chandrasekar Venkatraman, Ahmed Farahat, Teppei Inoue, Haiyan Wang, and Chetan Gupta. Both reviewed 2026 papers carry this researcher as first author without a corresponding-author flag, consistent with a compact industrial research team at Hitachi rather than a wide external network.
Activity77/100
Activity is the strongest axis at 77. OpenAlex counts by year show growth from 1 work each in 2017 and 2018 to a sustained 4–7 works annually from 2020 through 2026, peaking at 7 in 2022, with 13 works in the most recent three-year window. Four works are already recorded for 2026, indicating a research pace that is both steady and currently active.
Reproducibility60/100
Reproducibility scores 60, computed as the mean (60) over 2 Evolve-assessed papers. The two verdicts diverge: the 2026 planning-versus-RL multi-asset bearing maintenance benchmark is rated high readiness — honest and reproducible, though small in scale — while the hub-and-spoke agentic LLM supply chain framework is rated low, capped by a small synthetic testbed and no significance tests.
Research summary
Xian Yeow Lee is an industrial machine learning researcher at Hitachi in Japan whose public bibliographic record (OpenAlex, ORCID 0000-0001-6639-8398) shows steady growth: single works in 2017 and 2018 rising to a consistent four to seven publications per year from 2020 through 2026, with 13 works in the most recent three-year window, an h-index of 12 and an i10-index of 15. Topic modelling of that record clusters on reinforcement learning in robotics (11 works), adversarial robustness in machine learning (6), industrial vision systems and defect detection (6), advanced neural network applications (5), and machine fault diagnosis (5). Evolve's field distribution splits evenly (0.5/0.5) between reinforcement learning for predictive maintenance (PHM) and LLM-based multi-agent systems for enterprise supply chain analytics, and this dual track is visible in two 2026 first-author arXiv papers: a controlled planning-versus-RL benchmark for multi-asset bearing maintenance, rated high for reproducibility readiness, and a hub-and-spoke agentic LLM framework for supply chain analytics, rated low. Radar dimensions show strong activity (77), mid-range reproducibility (60 across two assessed papers), modest collaboration (36, with five distinct recent coauthors), and a focus score of 0 that matches the even two-field split; output and impact axes lack sufficient data and are left unscored.
Collaboration pattern
Collaboration is compact: 13 works in the recent three-year window with only 5 distinct coauthors (radar score 36). The recurring circle — Chandrasekar Venkatraman, Ahmed Farahat, Teppei Inoue, Haiyan Wang, and Chetan Gupta — suggests a stable core team, likely within Hitachi's industrial research group, and both reviewed 2026 papers carry this researcher as first author without a corresponding-author flag.
Research keywords
Highlights
First-author benchmark: planning vs. reinforcement learning in multi-asset maintenance
Agentic LLM framework for supply chain analytics
A dual-track industrial AI agenda at Hitachi
Evolve's even 0.5/0.5 field split and the OpenAlex topic counts — reinforcement learning in robotics (11), industrial vision and defect detection (6), machine fault diagnosis (5) — position this researcher across reliability engineering and enterprise AI, both exercised from an industrial laboratory setting with a small core team.
Timeline
- 2017First works appear in the OpenAlex record, with 1 work recorded in 2017 and 1 in 2018.
- 2020Output ramps up to 6 works and holds at 4–7 per year through 2026, marking the shift to a sustained research cadence.
- 2022Peak year in the record with 7 works published.
- 2026Two first-author arXiv papers enter the Evolve catalog: a planning-versus-RL multi-asset maintenance benchmark and a hub-and-spoke multi-agent LLM supply chain framework.
Papers in the Evolve catalog
Frequent co-authors
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.
- Bibliometric metrics are derived from OpenAlex only; the output and impact radar axes lack sufficient underlying data and are not scored.
- Evolve has reviewed 2 papers by this researcher; verdicts and reproducibility readiness apply only to those works.
- The reproducibility score (60) is a mean over 2 assessed papers and may shift as more works are reviewed.
- Identity is name-resolved from public bibliographic data (ORCID 0000-0001-6639-8398); biography beyond the stated affiliation is not covered.