{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T20:01:38Z","timestamp":1773000098515,"version":"3.50.1"},"reference-count":55,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T00:00:00Z","timestamp":1772150400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T00:00:00Z","timestamp":1772150400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Concurrency and Computation"],"published-print":{"date-parts":[[2026,3]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>Designing adaptive and energy\u2010aware scheduling algorithms for scientific workflows has become increasingly important as high\u2010performance computing (HPC) systems evolve toward large\u2010scale heterogeneous architectures. This paper proposes PPO\u2013KAN, a hybrid workflow scheduling framework that integrates proximal policy optimization (PPO) with Kolmogorov\u2013Arnold Networks (KAN) to jointly optimize makespan and energy consumption. PPO enables adaptive policy learning in dynamic resource environments, while the KAN\u2010based policy representation enhances expressiveness by modeling complex nonlinear task\u2013resource relationships. The proposed framework supports three reward formulations\u2014makespan\u2010oriented, energy\u2010oriented, and weighted multi\u2010objective\u2014allowing explicit control over optimization trade\u2010offs. Extensive experiments conducted on four benchmark scientific workflows\u2014Epigenomics (904 tasks), LIGO (922 tasks), Montage (902 tasks), and SIPHT (1004 tasks)\u2014using simulated heterogeneous HPC clusters demonstrate that PPO\u2013KAN achieves up to a 41.7% reduction in makespan and a 28.3% reduction in energy consumption compared with NSGA\u2010II and PPO with feedforward neural networks. Furthermore, convergence analysis shows faster and more stable learning, with policies stabilizing within 50\u2013100 episodes. Ablation studies further confirm that architectural enhancements improve energy efficiency and policy robustness. Overall, the results indicate that PPO\u2013KAN provides an effective and scalable solution for multi\u2010objective workflow scheduling in heterogeneous HPC environments.<\/jats:p>","DOI":"10.1002\/cpe.70624","type":"journal-article","created":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T14:24:47Z","timestamp":1772979887000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Energy\u2010Aware Multi\u2010Objective Workflow Scheduling Using Proximal Policy Optimization and Kolmogorov\u2013Arnold Networks in Heterogeneous\n                    <scp>HPC<\/scp>\n                    Environments"],"prefix":"10.1002","volume":"38","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0817-7009","authenticated-orcid":false,"given":"Sumit Kumar","family":"Saurav","sequence":"first","affiliation":[{"name":"Indian Institute of Information Technology Kottayam  Kottayam Kerala India"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shajulin","family":"Benedict","sequence":"additional","affiliation":[{"name":"Indian Institute of Information Technology Kottayam  Kottayam Kerala India"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2026,2,27]]},"reference":[{"key":"e_1_2_10_2_1","unstructured":"R.Ferreira Da Silva D.Bard K.Chard et al. \u201cWorkflows Community Summit 2024: Future Trends and Challenges in Scientific Workflows \u201d Technical Report (Oak Ridge Leadership Computing Facility (OLCF) Oak Ridge National Laboratory (ORNL) 2024)."},{"issue":"8","key":"e_1_2_10_3_1","first-page":"129","article-title":"Integrating HPC, AI, and Workflows for Scientific Data Analysis","volume":"13","author":"Badia R. M.","year":"2024","journal-title":"Dagstuhl Reports"},{"key":"e_1_2_10_4_1","unstructured":"\u201cTop 500 Supercomputers \u201d accessed December 2024 https:\/\/top500.org."},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1177\/1094342017704893"},{"key":"e_1_2_10_6_1","unstructured":"R. F.daSilva H.Casanova K.Chard et al. \u201cA Community Roadmap for Scientific Workflows Research and Development \u201d in2021 IEEE Workshop on Workflows in Support of Large\u2010Scale Science (WORKS) St. Louis MO USA (2021) 81\u201390."},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.rser.2023.114019"},{"key":"e_1_2_10_8_1","doi-asserted-by":"crossref","unstructured":"S. K.SauravandS.Benedict \u201cA Taxonomy and Survey on Energy\u2010Aware Scientific Workflows Scheduling in Large\u2010Scale Heterogeneous Architecture \u201d in2021 6th International Conference on Inventive Computation Technologies (ICICT)(2021) 820\u2013826.","DOI":"10.1109\/ICICT50816.2021.9358707"},{"key":"e_1_2_10_9_1","doi-asserted-by":"crossref","unstructured":"S. K.SauravandS.Benedict \u201cEnergy Aware Scheduling Algorithms for Cloud Environments\u00a0\u2013\u00a0A Survey \u201d in2021 2nd International Conference on Advances in Computing Communication Embedded and Secure Systems (ACCESS)(2021) 50\u201355.","DOI":"10.1109\/ACCESS51619.2021.9563340"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/71.993206"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2022.108791"},{"issue":"1","key":"e_1_2_10_12_1","first-page":"1","article-title":"A Comprehensive Review and Evaluation of Permutation Flowshop Heuristics","volume":"208","author":"Pan Q.","year":"2011","journal-title":"European Journal of Operational Research"},{"key":"e_1_2_10_13_1","doi-asserted-by":"crossref","unstructured":"H.Mao M.Schwarzkopf S. B.Venkatakrishnan Z.Meng andM.Alizadeh \u201cLearning Scheduling Algorithms for Data Processing Clusters \u201d inACM SIGCOMM(2019).","DOI":"10.1145\/3341302.3342080"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10115-024-02167-7"},{"key":"e_1_2_10_15_1","unstructured":"J.Schulman F.Wolski P.Dhariwal A.Radford andO.Klimov \u201cProximal Policy Optimization Algorithms \u201d preprint arXiv:1707.06347 2017."},{"key":"e_1_2_10_16_1","doi-asserted-by":"crossref","unstructured":"A.Jayanetti S.Halgamuge andR.Buyya \u201cReinforcement Learning Based Workflow Scheduling in Cloud and Edge Computing Environments: A Taxonomy Review and Future Directions \u201d preprint arXiv:2408.02938 2023.","DOI":"10.1145\/3685767.3685780"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3500046"},{"key":"e_1_2_10_18_1","unstructured":"X.Chen Z.Liu Z.Li J.Zhang S.Boyd andJ. Z.Kolter \u201cKolmogorov\u2013Arnold Networks for Scientific Machine Learning \u201d inAdvances in Neural Information Processing Systems (NeurIPS)(Curran Associates Inc. 2023)."},{"key":"e_1_2_10_19_1","doi-asserted-by":"crossref","unstructured":"N.Le A. P.Ngo andH. T.Nguyen \u201cKolmogorov\u2013Arnold Networks for Supervised Learning in Smart Grids \u201d inNorth American Power Symposium (NAPS)(2024).","DOI":"10.1109\/NAPS61145.2024.10741767"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3509218"},{"issue":"4","key":"e_1_2_10_21_1","first-page":"1311","article-title":"A Performance\u2010Effective Task Scheduling Algorithm for Heterogeneous Computing","volume":"71","author":"Mei L.","year":"2015","journal-title":"Journal of Supercomputing"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/71.503776"},{"issue":"2","key":"e_1_2_10_23_1","first-page":"94","article-title":"Low Complexity Performance Effective Task Scheduling Algorithm for Heterogeneous Computing Environments","volume":"3","author":"Ilavarasan E.","year":"2007","journal-title":"Journal of Computer Sciences"},{"key":"e_1_2_10_24_1","doi-asserted-by":"crossref","unstructured":"H.Mao M.Alizadeh I.Menache andS.Kandula \u201cResource Management with Deep Reinforcement Learning \u201d inProceedings of the 15th ACM Workshop on Hot Topics in Networks (HotNets)(ACM 2016) 50\u201356.","DOI":"10.1145\/3005745.3005750"},{"key":"e_1_2_10_25_1","first-page":"62","article-title":"Reinforcement Learning\u2010Based Workflow Scheduling in Heterogeneous Computing Systems","volume":"175","author":"Zhang Y.","year":"2023","journal-title":"Journal of Parallel and Distributed Computing"},{"key":"e_1_2_10_26_1","unstructured":"X.Ren R.Mahmud andR.Buyya \u201cA Deep Reinforcement Learning Scheduler for HPC Workloads \u201d in2019 IEEE 27th International Symposium on Modeling Analysis and Simulation of Computer and Telecommunication Systems (MASCOTS)(IEEE 2019) 29\u201342."},{"key":"e_1_2_10_27_1","doi-asserted-by":"crossref","unstructured":"G. P.Rodrigo E.Elmroth P.\u2010O.\u00d6stberg andL.Ramakrishnan \u201cEnabling Workflow\u2010Aware Scheduling on HPC Systems \u201d inHPDC'17: Proceedings of the 26th International Symposium on High\u2010Performance Parallel and Distributed Computing(2017) 3\u201314 https:\/\/doi.org\/10.1145\/3078597.3078604.","DOI":"10.1145\/3078597.3078604"},{"key":"e_1_2_10_28_1","first-page":"319","article-title":"Different Aspects of Workflow Scheduling in Large\u2010Scale Distributed Systems","volume":"67","author":"Stavrinides G. L.","year":"2017","journal-title":"Future Generation Computer Systems"},{"key":"e_1_2_10_29_1","first-page":"192","article-title":"Hierarchical Scheduling Strategies for Large\u2010Scale Distributed Workflow Execution","volume":"37","author":"Rodriguez M.","year":"2014","journal-title":"Future Generation Computer Systems"},{"key":"e_1_2_10_30_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10586-021-03432-y"},{"key":"e_1_2_10_31_1","doi-asserted-by":"publisher","DOI":"10.3390\/app10186611"},{"key":"e_1_2_10_32_1","first-page":"1","article-title":"A Workflow Scheduling Method for Cloud Computing Platform","volume":"122","author":"Rajak R.","year":"2022","journal-title":"Wireless Personal Communications"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1109\/4235.996017"},{"key":"e_1_2_10_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/TEVC.2004.826067"},{"key":"e_1_2_10_35_1","doi-asserted-by":"publisher","DOI":"10.1504\/IJOR.2019.101456"},{"key":"e_1_2_10_36_1","doi-asserted-by":"publisher","DOI":"10.4018\/IJAL.2019070102"},{"key":"e_1_2_10_37_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2024.111366"},{"key":"e_1_2_10_38_1","doi-asserted-by":"publisher","DOI":"10.1080\/01605682.2023.2195426"},{"key":"e_1_2_10_39_1","doi-asserted-by":"crossref","unstructured":"S. G.Machicao\u2010MollocondoandE.Hinojosa\u2010Cardenas \u201cParallel Multi\u2010Objective Evolutionary Algorithms: A Systematic Literature Review \u201d inInformation Management and Big Data. SIMBig 2024. Communications in Computer and Information Science vol 2496 (2025) 93\u2013109.","DOI":"10.1007\/978-3-031-91428-7_7"},{"key":"e_1_2_10_40_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.sysarc.2020.101837"},{"key":"e_1_2_10_41_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10586-021-03269-5"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10586-020-03085-3"},{"key":"e_1_2_10_43_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10766-021-00721-2"},{"key":"e_1_2_10_44_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCASAI.2024.3476237"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2022.11.032"},{"key":"e_1_2_10_46_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.compeleceng.2023.108653"},{"key":"e_1_2_10_47_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2022.109707"},{"key":"e_1_2_10_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2024.3524506"},{"key":"e_1_2_10_49_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00500-020-04931-7"},{"key":"e_1_2_10_50_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10586-024-04828-2"},{"key":"e_1_2_10_51_1","unstructured":"J.Luo Y.Zhou X.Li M.Yuan J.Yao andJ.Zeng \u201cLearning to Optimize DAG Scheduling in Heterogeneous Environment \u201d preprint arXiv:2103.06980 2021 https:\/\/doi.org\/10.48550\/arXiv.2103.06980."},{"key":"e_1_2_10_52_1","doi-asserted-by":"publisher","DOI":"10.15587\/1729\u20104061.2025.329263"},{"key":"e_1_2_10_53_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11227-025-07685-x"},{"issue":"1","key":"e_1_2_10_54_1","first-page":"362","article-title":"MOHEFT: A Multi\u2010Objective List\u2010Based Method for Workflow Scheduling","volume":"29","author":"Durillo J. J.","year":"2012","journal-title":"Future Generation Computer Systems"},{"key":"e_1_2_10_55_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.future.2014.10.008"},{"key":"e_1_2_10_56_1","doi-asserted-by":"publisher","DOI":"10.1007\/s42514\u2010021\u201000083\u20108"}],"container-title":["Concurrency and Computation: Practice and Experience"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpe.70624","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/cpe.70624","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/cpe.70624","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,8]],"date-time":"2026-03-08T17:30:56Z","timestamp":1772991056000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/cpe.70624"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,27]]},"references-count":55,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,3]]}},"alternative-id":["10.1002\/cpe.70624"],"URL":"https:\/\/doi.org\/10.1002\/cpe.70624","archive":["Portico"],"relation":{},"ISSN":["1532-0626","1532-0634"],"issn-type":[{"value":"1532-0626","type":"print"},{"value":"1532-0634","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,27]]},"assertion":[{"value":"2025-08-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-09","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-02-27","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70624"}}