{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T12:34:19Z","timestamp":1775219659021,"version":"3.50.1"},"reference-count":33,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,12,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Chemical process optimization is crucial to maximize production efficiency and economic performance. Optimization algorithms, including gradient-based solvers, numerical methods, and parameter grid searches, become impractical when operating constraints are ill-defined or unavailable, requiring engineers to rely on subjective heuristics to estimate feasible parameter ranges. To address this constraint definition bottleneck, we present a multi-agent framework of large language model (LLM) agents that autonomously infer operating constraints from minimal process descriptions, then collaboratively guide optimization using the inferred constraints. Our AutoGen-based agentic framework employs OpenAI\u2019s o3 model, with specialized agents for constraint generation, parameter validation, simulation execution, and optimization guidance. Through two phases: (i) autonomous constraint generation using embedded domain knowledge, and (ii) iterative multi-agent optimization, the framework eliminates the need for predefined operational bounds. Validated on the hydrodealkylation process across cost, yield, and yield-to-cost ratio metrics, the framework demonstrated competitive performance with conventional optimization methods while achieving a 31-fold reduction in wall-time relative to grid search, converging in under 20 min and requiring far fewer iterations to converge. Beyond computational efficiency, the framework\u2019s reasoning-guided search demonstrates sophisticated process understanding, correctly identifying utility trade-offs, and applying domain-informed heuristics. Unlike conventional optimization methods like Bayesian optimization that require predefined constraints, our approach uniquely combines autonomous constraint generation with interpretable, reasoning-guided parameter exploration. Reproducibility analysis across five independent trials demonstrates consistent convergence behavior, while model comparison reveals that reasoning-capable LLM architectures (o3, o1) are essential for successful optimization, with standard models failing to converge effectively. This approach shows significant potential for optimization scenarios where operational constraints are poorly characterized or unavailable, particularly for emerging processes and retrofit applications.<\/jats:p>","DOI":"10.1088\/2632-2153\/ae2382","type":"journal-article","created":{"date-parts":[[2025,11,24]],"date-time":"2025-11-24T22:53:38Z","timestamp":1764024818000},"page":"045067","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["LLM-guided chemical process optimization with a multi-agent approach"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-7416-8522","authenticated-orcid":true,"given":"Tong","family":"Zeng","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0004-6517-3145","authenticated-orcid":true,"given":"Srivathsan","family":"Badrinarayanan","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0000-0370-4212","authenticated-orcid":true,"given":"Janghoon","family":"Ock","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0009-2118-2935","authenticated-orcid":true,"given":"Cheng-Kai","family":"Lai","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2952-8576","authenticated-orcid":true,"given":"Amir","family":"Barati Farimani","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,12,15]]},"reference":[{"key":"mlstae2382bib1","author":"Biegler","year":"2010","type":"book"},{"key":"mlstae2382bib2","first-page":"2951","type":"conference-proceedings","article-title":"Practical Bayesian optimization of machine learning algorithms","author":"Snoek","year":"2012"},{"key":"mlstae2382bib3","doi-asserted-by":"publisher","DOI":"10.1016\/j.coche.2021.100728","type":"journal-article","article-title":"Bayesian optimization for chemical products and functional materials","volume":"36","author":"Wang","year":"2022","journal-title":"Curr. Opin. Chem. Eng."},{"key":"mlstae2382bib4","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1007\/s10107-004-0559-y","type":"journal-article","article-title":"On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming","volume":"106","author":"W\u00e4chter","year":"2006","journal-title":"Math. Program."},{"key":"mlstae2382bib5","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1137\/S0036144504446096","type":"journal-article","article-title":"SNOPT: an SQP algorithm for large-scale constrained optimization","volume":"47","author":"Gill","year":"2005","journal-title":"SIAM Rev."},{"key":"mlstae2382bib6","doi-asserted-by":"publisher","first-page":"1169","DOI":"10.1016\/j.compchemeng.2003.11.003","type":"journal-article","article-title":"Retrospective on optimization","volume":"28","author":"Biegler","year":"2004","journal-title":"Comput. Chem. Eng."},{"key":"mlstae2382bib7","doi-asserted-by":"publisher","first-page":"S229","DOI":"10.1016\/S0098-1354(98)00059-3","type":"journal-article","article-title":"Total process optimization in chemical engineering with evolutionary algorithms","volume":"22","author":"Gross","year":"1998","journal-title":"Comput. Chem. Eng."},{"key":"mlstae2382bib8","doi-asserted-by":"publisher","first-page":"1247","DOI":"10.1007\/s10898-012-9951-y","type":"journal-article","article-title":"Derivative-free optimization: a review of algorithms and comparison of software implementations","volume":"56","author":"Rios","year":"2013","journal-title":"J. Glob. Optim."},{"key":"mlstae2382bib9","author":"Edgar","year":"2001","edition":"2nd edn","type":"book"},{"key":"mlstae2382bib10","author":"James","year":"1988","type":"book"},{"key":"mlstae2382bib11","author":"Himmelblau","year":"2005","edition":"subsequent edn","type":"book"},{"key":"mlstae2382bib12","doi-asserted-by":"publisher","DOI":"10.1002\/amp2.10095","type":"journal-article","article-title":"The IDAES process modeling framework and model library\u2013flexibility for process simulation and optimization","volume":"3","author":"Lee","year":"2021","journal-title":"J. Adv. Manuf. Process."},{"key":"mlstae2382bib13","author":"Bynum","year":"2021","edition":"3rd edn","type":"book"},{"key":"mlstae2382bib14","first-page":"24824","type":"conference-proceedings","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"vol 35","author":"Wei","year":"2022"},{"key":"mlstae2382bib15","first-page":"1877","type":"conference-proceedings","article-title":"Language models are few-shot learners","volume":"vol 33","author":"Brown","year":"2020"},{"key":"mlstae2382bib16","doi-asserted-by":"publisher","first-page":"1373","DOI":"10.1038\/s42256-025-01110-x","type":"journal-article","article-title":"Towards agentic science for advancing scientific discovery","volume":"7","author":"Xin","year":"2025","journal-title":"Nat. Mach. Intell."},{"key":"mlstae2382bib17","first-page":"uyTL5Bvosj","type":"preprint","article-title":"Beyond the imitation game: quantifying and extrapolating the capabilities of language models","author":"Srivastava","year":"2023"},{"key":"mlstae2382bib18","article-title":"The AI scientist: towards fully automated open-ended scientific discovery","author":"Chris","year":"2024","type":"preprint"},{"key":"mlstae2382bib19","article-title":"Adsorb-agent: autonomous identification of stable adsorption configurations via large language model agent","author":"Ock","year":"2024","type":"preprint"},{"key":"mlstae2382bib20","doi-asserted-by":"publisher","DOI":"10.26434\/chemrxiv-2025-zkn81-v2","type":"preprint","article-title":"Modular large language model agents for multi-task computational materials science","author":"Chaudhari","year":"2025"},{"key":"mlstae2382bib21","doi-asserted-by":"publisher","DOI":"10.1016\/j.addma.2025.105027","type":"journal-article","article-title":"LLM-3D print: large language models to monitor and control 3D printing","volume":"114","author":"Jadhav","year":"2025","journal-title":"Addit. Manuf."},{"key":"mlstae2382bib22","article-title":"Chemical reasoning in LLMs unlocks steerable synthesis planning and reaction mechanism elucidation","author":"Bran","year":"2025","type":"preprint"},{"key":"mlstae2382bib23","doi-asserted-by":"publisher","first-page":"525","DOI":"10.1038\/s42256-024-00832-8","type":"journal-article","article-title":"Augmenting large language models with chemistry tools","volume":"6","author":"Bran","year":"2024","journal-title":"Nat. Mach. Intell."},{"key":"mlstae2382bib24","article-title":"Controlagent: automating control system design via novel integration of llm agents and domain expertise","author":"Guo","year":"2024","type":"preprint"},{"key":"mlstae2382bib25","article-title":"Autonomous industrial control using an agentic framework with large language models","author":"Vyas","year":"2024","type":"preprint"},{"key":"mlstae2382bib26","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1038\/s44286-024-00041-5","type":"journal-article","article-title":"Generative artificial intelligence in chemical engineering","volume":"1","author":"Schweidtmann","year":"2024","journal-title":"Nat. Chem. Eng."},{"key":"mlstae2382bib27","doi-asserted-by":"publisher","first-page":"3109","DOI":"10.1016\/B978-0-443-28824-1.50519-6","type":"conference-proceedings","article-title":"Toward autocorrection of chemical process flowsheets using large language models","author":"Balhorn","year":"2024"},{"key":"mlstae2382bib28","doi-asserted-by":"publisher","DOI":"10.1002\/aic.18259","type":"journal-article","article-title":"Toward automatic generation of control structures for process flow diagrams with large language models","volume":"70","author":"Hirtreiter","year":"2024","journal-title":"AIChE J."},{"key":"mlstae2382bib29","article-title":"Autogen: enabling next-gen LLM applications via multi-agent conversation","author":"Wu","year":"2024","type":"conference-proceedings"},{"key":"mlstae2382bib30","article-title":"Large language model","author":"OpenAI O3 and o4-mini system card","year":"2025","type":"other"},{"key":"mlstae2382bib31","author":"Zeng","year":"2024","type":"book"},{"key":"mlstae2382bib32","article-title":"HDA flowsheet with distillation: adding expressions to compute capital and operating costs","author":"IDAES Team","year":"2024","type":"other"},{"key":"mlstae2382bib33","article-title":"Selective steam dealkylation of alkyl aromatic hydrocarbons","author":"Abdellatif","year":"2008","type":"other"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T11:07:52Z","timestamp":1765796872000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ae2382"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,15]]},"references-count":33,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2025,12,15]]},"published-print":{"date-parts":[[2025,12,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ae2382","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,15]]},"assertion":[{"value":"LLM-guided chemical process optimization with a multi-agent approach","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-07-02","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-11-24","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-12-15","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}