{"id":2330,"date":"2018-03-01T18:46:44","date_gmt":"2018-03-01T18:46:44","guid":{"rendered":"http:\/\/docs.creativegigs.net\/docs\/gullu-wp\/solved-issues\/your-theme-gullu-contains-outdated-copies-of-some-woocommerce-template-files\/"},"modified":"2025-12-15T19:17:46","modified_gmt":"2025-12-15T19:17:46","slug":"your-theme-gullu-contains-outdated-copies-of-some-woocommerce-template-files","status":"publish","type":"docs","link":"https:\/\/sahelib.atatec-design.com\/index.php\/docs\/gullu-knowledge-base\/solved-issues\/your-theme-gullu-contains-outdated-copies-of-some-woocommerce-template-files\/","title":{"rendered":"Apprentissage par renforcement dans la gestion des risques des projets de construction pharmaceutique"},"content":{"rendered":"\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\">\n<h2 class=\"wp-block-heading\">Points forts<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Nous avons examin\u00e9 86 \u00e9tudes sur l&#8217;apprentissage par renforcement (RL) dans les projets de construction pharmaceutique entre 2016 et 2025.<\/li>\n\n\n\n<li>Identification des types d&#8217;algorithmes d&#8217;apprentissage par renforcement et des sujets de recherche les plus influents.<\/li>\n\n\n\n<li>Nous avons d\u00e9crit les principes, les avantages et les inconv\u00e9nients de six algorithmes d&#8217;apprentissage par renforcement.<\/li>\n\n\n\n<li>Mise en lumi\u00e8re des principaux d\u00e9fis auxquels est confront\u00e9e la recherche en apprentissage automatique dans les projets de construction pharmaceutique.<\/li>\n\n\n\n<li>Strat\u00e9gies de diffusion sp\u00e9cifiques et orientations futures de la recherche indiqu\u00e9es.<\/li>\n<\/ul>\n<\/blockquote>\n\n\n\n<h2 class=\"wp-block-heading\">R\u00e9sum\u00e9s<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">La construction intelligente d&#8217;installations pharmaceutiques est confront\u00e9e \u00e0 des risques dynamiques et non lin\u00e9aires, et les m\u00e9thodes de gestion traditionnelles peinent \u00e0 r\u00e9pondre aux exigences \u00e9lev\u00e9es de r\u00e9activit\u00e9 et de conformit\u00e9 en temps r\u00e9el. Or, les recherches existantes sur l&#8217;apprentissage par renforcement (AR) dans ce domaine manquent encore d&#8217;une architecture d&#8217;application syst\u00e9matique et de consid\u00e9rations relatives \u00e0 la gouvernance industrielle. Par cons\u00e9quent, cet article examine les applications pratiques de six algorithmes \u2013 Deep Q-Network (DQN), Deep Deterministic Policy Gradient (DDPG) et Proximity Policy Optimization (PPO) \u2013 dans les domaines de la s\u00e9curit\u00e9 de la construction, du contr\u00f4le de la temp\u00e9rature, de la planification des ressources et de l&#8217;optimisation automatis\u00e9e des \u00e9quipements, validant ainsi le potentiel de l&#8217;apprentissage par renforcement pour g\u00e9rer efficacement les risques dynamiques gr\u00e2ce \u00e0 un apprentissage adaptatif. Parall\u00e8lement, cet article identifie avec pr\u00e9cision les principaux obstacles rencontr\u00e9s dans les applications actuelles&nbsp;: l&#8217;\u00e9cart de fid\u00e9lit\u00e9 entre l&#8217;environnement de simulation et la r\u00e9glementation m\u00e9dicale r\u00e9elle, l&#8217;absence de proc\u00e9dures de d\u00e9ploiement standardis\u00e9es pour l&#8217;apprentissage par renforcement et l&#8217;ambigu\u00eft\u00e9 entre l&#8217;autorit\u00e9 de d\u00e9cision algorithmique et la responsabilit\u00e9 de supervision humaine. Pour rem\u00e9dier \u00e0 ces probl\u00e8mes, cet article pr\u00e9sente un syst\u00e8me de simulation d&#8217;environnement haute fid\u00e9lit\u00e9 int\u00e9grant de multiples technologies, un cadre d&#8217;application d&#8217;apprentissage par renforcement standardis\u00e9 et un syst\u00e8me de gouvernance clair des droits et responsabilit\u00e9s, fournissant un soutien th\u00e9orique crucial et des voies pratiques pour la construction d&#8217;un paradigme fiable et efficace de gestion des risques li\u00e9s \u00e0 la construction d&#8217;installations pharmaceutiques.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">R\u00e9sum\u00e9 graphique<\/h2>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img fetchpriority=\"high\" decoding=\"async\" width=\"1024\" height=\"458\" src=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-1024x458.png\" alt=\"\" class=\"wp-image-6654\" style=\"aspect-ratio:2.2355638030782665;width:592px;height:auto\" srcset=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-1024x458.png 1024w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-300x134.png 300w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-768x344.png 768w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-20x9.png 20w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-32x14.png 32w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57-600x268.png 600w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-15.43.57.png 1082w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<h2 class=\"wp-block-heading\">Mots cl\u00e9s<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Apprentissage par renforcement<\/li>\n\n\n\n<li>projets de construction pharmaceutiques<\/li>\n\n\n\n<li>Gestion des risques<\/li>\n\n\n\n<li>Gradient de politique d\u00e9terministe profond<\/li>\n\n\n\n<li>Optimisation de la politique proximale<\/li>\n\n\n\n<li>R\u00e9seau Q profond<\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"cesectitle0006\">1.&nbsp;Introduction\u200b<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">L&#8217;industrie pharmaceutique est un secteur vital de l&#8217;\u00e9conomie nationale, du bien-\u00eatre de la population et de la s\u00e9curit\u00e9 nationale. Selon les derni\u00e8res donn\u00e9es du minist\u00e8re de l&#8217;Industrie et des Technologies de l&#8217;information, depuis le d\u00e9but du 14e plan quinquennal, l&#8217;industrie pharmaceutique chinoise a enregistr\u00e9 un taux de croissance annuel moyen de 9,3 % pour son chiffre d&#8217;affaires principal, de 11,3 % pour ses b\u00e9n\u00e9fices totaux et de plus de 20 % pour ses investissements en R&amp;D [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0001\">1<\/a>\u00a0]. Les projets de construction d&#8217;installations pharmaceutiques constituent le socle mat\u00e9riel du d\u00e9veloppement durable de l&#8217;industrie, garantissant la s\u00e9curit\u00e9 d&#8217;approvisionnement en m\u00e9dicaments et permettant l&#8217;it\u00e9ration et la modernisation technologiques. Ces projets pr\u00e9sentent souvent des caract\u00e9ristiques distinctes dans leurs processus et m\u00e9thodologies de mise en \u0153uvre [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0002\">2<\/a>\u00a0]. Contrairement aux projets de construction classiques, leur objectif principal d\u00e9passe la simple construction d&#8217;un espace physique\u00a0; il s&#8217;agit fondamentalement d&#8217;\u00e9tablir un environnement de production propre, hautement contr\u00f4l\u00e9, v\u00e9rifiable et conforme aux BPF (Bonnes Pratiques de Fabrication) [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0003\">3<\/a>\u00a0]. Par cons\u00e9quent, elle d\u00e9passe largement le cadre de la construction et de l&#8217;installation de b\u00e2timents conventionnels, int\u00e9grant pleinement des technologies complexes de salles blanches (telles que les syst\u00e8mes CVC et les syst\u00e8mes d&#8217;eau purifi\u00e9e), des sch\u00e9mas de flux de processus rigoureux et des activit\u00e9s de validation exhaustives. Sa gestion des risques est constamment confront\u00e9e \u00e0 des d\u00e9fis consid\u00e9rables pos\u00e9s par des environnements dynamiques, de multiples contraintes et des \u00e9v\u00e9nements impr\u00e9vus, m\u00eame rares. Les approches traditionnelles de gestion des risques, souvent fond\u00e9es sur une exp\u00e9rience statique et des r\u00e8gles pr\u00e9d\u00e9finies, sont fr\u00e9quemment inad\u00e9quates et rigides face aux incertitudes constantes qui surgissent tout au long du cycle de vie d&#8217;un projet [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0004\">4<\/a>\u00a0,\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0005\">5<\/a>\u00a0]. Par exemple, les approches traditionnelles d&#8217;apprentissage automatique (AA) se r\u00e9v\u00e8lent souvent inadapt\u00e9es aux sc\u00e9narios de risques dynamiques, multidimensionnels et non lin\u00e9aires en raison de leur capacit\u00e9 limit\u00e9e d&#8217;interaction soutenue avec l&#8217;environnement [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0006\">6<\/a>\u00a0,\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0007\">7<\/a>\u00a0].Ces derni\u00e8res ann\u00e9es, les avanc\u00e9es en intelligence artificielle (IA), et notamment en apprentissage par renforcement (AR), avec son paradigme central distinctif d\u2019\u00ab\u00a0agents apprenant de mani\u00e8re autonome des strat\u00e9gies optimales gr\u00e2ce \u00e0 l\u2019interaction avec leur environnement\u00a0\u00bb, ont ouvert de nouvelles perspectives pour relever ce d\u00e9fi [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0008\">8<\/a>\u00a0]. L\u2019AR, en tant que classe de techniques de prise de d\u00e9cision autonomes bas\u00e9es sur le processus de d\u00e9cision markovien, optimise les strat\u00e9gies dans des environnements dynamiques gr\u00e2ce au m\u00e9canisme d\u2019\u00ab\u00a0essais et erreurs avec r\u00e9troaction\u00a0\u00bb, offrant ainsi une approche in\u00e9dite pour r\u00e9soudre le probl\u00e8me multidimensionnel de la gestion intelligente des risques dans la construction [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0009\">9<\/a>\u00a0]. Compar\u00e9 aux m\u00e9thodes traditionnelles, l\u2019AR pr\u00e9sente des avantages uniques dans trois domaines. Premi\u00e8rement, il ne n\u00e9cessite pas la d\u00e9finition de r\u00e8gles heuristiques et d\u2019it\u00e9rations distinctes pour les autres t\u00e2ches de construction\u00a0; il apprend automatiquement diverses strat\u00e9gies d\u2019optimisation lors de l\u2019entra\u00eenement et de la simulation [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0056\">56<\/a>\u00a0]. Deuxi\u00e8mement, sa capacit\u00e9 de repr\u00e9sentation efficace des espaces d\u2019\u00e9tats de grande dimension a permis de r\u00e9duire consid\u00e9rablement la d\u00e9pendance aux connaissances et au jugement d\u2019experts dans des processus tels que la mod\u00e9lisation des informations du b\u00e2timent (BIM) [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0050\">50<\/a>\u00a0]. Plus important encore, son cadre offre une aide \u00e0 la d\u00e9cision distribu\u00e9e pour les syst\u00e8mes collaboratifs tels que les parcs de grues \u00e0 tour et les flottes de robots, en se caract\u00e9risant par l&#8217;auto-apprentissage, la robustesse et l&#8217;adaptabilit\u00e9 [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0162\">162<\/a>\u00a0]. Par exemple, Han et al. [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0010\">10<\/a>\u00a0] ont d\u00e9velopp\u00e9 une m\u00e9thode d\u00e9centralis\u00e9e de planification modulaire de la r\u00e9novation d&#8217;h\u00f4pitaux bas\u00e9e sur des algorithmes d&#8217;apprentissage par renforcement profond \u00e0 m\u00e9moire \u00e9tendue. Guerrero et al. [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0011\">11<\/a>\u00a0] ont d\u00e9velopp\u00e9 un syst\u00e8me d&#8217;aide \u00e0 la d\u00e9cision pour la conception de b\u00e2timents de sant\u00e9, bas\u00e9 sur le raisonnement \u00e0 partir de cas et l&#8217;apprentissage par renforcement, d\u00e9montrant une grande efficacit\u00e9 dans la d\u00e9tection des d\u00e9fauts et des erreurs. Ainsi, comme le montre la\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#fig0001\">figure 1<\/a>\u00a0, au cours de la derni\u00e8re d\u00e9cennie, le nombre d&#8217;articles sur l&#8217;apprentissage par renforcement a connu une croissance exponentielle, passant de 2\u00a0268 en 2015 \u00e0 35\u00a0488 en 2024. Cette tendance s&#8217;observe \u00e9galement pour le nombre annuel de publications sur l&#8217;apprentissage par renforcement dans le secteur de la construction et la gestion des risques.<\/p>\n\n\n\n\n\n<p class=\"has-text-align-center wp-block-paragraph\">Fig. 1.\u00a0Tendances\u00a0annuelles des publications en apprentissage par renforcement.<\/p>\n\n\n\n<p class=\"has-text-align-left wp-block-paragraph\">2.&nbsp;Principe de base de l&#8217;apprentissage par&nbsp;renforcement<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Comme indiqu\u00e9 sur\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#fig0002\">la figure 2<\/a>\u00a0, le processus de d\u00e9cision markovien (MDP) est au c\u0153ur de l&#8217;apprentissage par renforcement (RL). En RL, un agent apprend en interagissant avec son environnement, cherchant \u00e0 prendre des d\u00e9cisions bas\u00e9es sur l&#8217;\u00e9tat de ce dernier afin de maximiser les r\u00e9compenses \u00e0 long terme. Le MDP fournit un cadre pour mod\u00e9liser la relation entre l&#8217;intelligence et l&#8217;environnement dans les probl\u00e8mes d&#8217;apprentissage par renforcement, aidant ainsi \u00e0 d\u00e9crire comment prendre des d\u00e9cisions optimales dans des probl\u00e8mes de d\u00e9cision dynamiques et s\u00e9quentiels [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0082\">82<\/a>\u00a0]. Le MDP est compos\u00e9 d&#8217;un quintuplet.o\u00f9est l&#8217;espace d&#8217;\u00e9tat,est l&#8217;espace d&#8217;action,est la probabilit\u00e9 de transfert d&#8217;\u00e9tat,est la fonction de r\u00e9compense, et \u03b3 est le facteur d&#8217;actualisation. L&#8217;objectif est de trouver une politique optimalequi maximise la r\u00e9compense cumulative \u00e0 long terme et satisfait l&#8217;\u00e9quation de Bellman [\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0083\">83<\/a>\u00a0] :<br><\/p>\n\n\n\n\n\n<p class=\"has-text-align-center wp-block-paragraph\">Fig. 2.\u00a0Apprentissage\u00a0par renforcement avec processus de cha\u00eene de Markov.<\/p>\n\n\n\n<h2 class=\"wp-block-heading has-text-align-left\">3.\u00a0M\u00e9thodes\u00a0et mat\u00e9riaux<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"cesectitle0009\">3.1&nbsp;.&nbsp;M\u00e9thodes de recherche<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">La pr\u00e9sentation de cette revue syst\u00e9matique suit les crit\u00e8res de la d\u00e9claration PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses).  Elle vise \u00e0 aider les auteurs \u00e0 pr\u00e9senter de mani\u00e8re transparente la motivation, les m\u00e9thodes et les r\u00e9sultats de leurs revues syst\u00e9matiques, am\u00e9liorant ainsi la transparence et la reproductibilit\u00e9 de leurs \u00e9tudes. Par ailleurs, elle a mis en \u0153uvre une m\u00e9thodologie de recherche hybride combinant bibliom\u00e9trie et analyse de contenu.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">3.2&nbsp;.&nbsp;Mat\u00e9riel de recherche<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Le processus de recherche, de collecte et d&#8217;analyse des donn\u00e9es de cette \u00e9tude est r\u00e9sum\u00e9 dans\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#fig0003\">la figure 3.<\/a>\u00a0Premi\u00e8rement, les bases de donn\u00e9es Web of Science (WoS) et Scopus, reconnues par de nombreuses \u00e9tudes comme \u00e9tant parfaitement adapt\u00e9es \u00e0 l&#8217;analyse d&#8217;articles sur l&#8217;ing\u00e9nierie et les technologies innovantes, ont \u00e9t\u00e9 s\u00e9lectionn\u00e9es comme sources de recherche bibliographique. Deuxi\u00e8mement, comme indiqu\u00e9 dans\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#tbl0001\">le tableau 1<\/a>\u00a0, des mots-cl\u00e9s tels que \u00ab\u00a0Installation pharmaceutique\u00a0\u00bb, \u00ab\u00a0Risque de s\u00e9curit\u00e9\u00a0\u00bb et \u00ab\u00a0Apprentissage par renforcement Q\u00a0\u00bb ont \u00e9t\u00e9 utilis\u00e9s. Les codes de recherche sp\u00e9cifiques \u00e0 chaque base de donn\u00e9es sont list\u00e9s dans le\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#tbl0002\">tableau 2.<\/a>\u00a0Apr\u00e8s une premi\u00e8re s\u00e9lection, 75 r\u00e9f\u00e9rences ont \u00e9t\u00e9 obtenues de WoS et 98 de Scopus. Ces articles ont ensuite \u00e9t\u00e9 examin\u00e9s afin d&#8217;\u00e9liminer les r\u00e9f\u00e9rences non pertinentes et les doublons, sur la base des titres et des r\u00e9sum\u00e9s, ce qui a permis de retenir 106 documents. Troisi\u00e8mement, le texte int\u00e9gral a \u00e9t\u00e9 \u00e9valu\u00e9 selon les crit\u00e8res suivants\u00a0: (1) la gestion des risques devait \u00eatre directement ou indirectement li\u00e9e \u00e0 la construction\u00a0; (2) les \u00e9tudes ne pr\u00e9sentant qu&#8217;une br\u00e8ve introduction \u00e0 l&#8217;apprentissage par renforcement Q ont \u00e9t\u00e9 exclues. (3) Conform\u00e9ment \u00e0 la norme ISO 31000:2018, les \u00e9tudes ne comportant aucune \u00e9tape de gestion des risques (identification, analyse, \u00e9valuation et suivi) doivent \u00eatre exclues. Au final, 86 documents ont \u00e9t\u00e9 retenus pour l\u2019analyse ult\u00e9rieure.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Tableau 1.&nbsp;Mots&nbsp;-cl\u00e9s utilis\u00e9s pour la recherche bibliographique.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th class=\"has-text-align-left\" data-align=\"left\">Domaine de recherche<\/th><th class=\"has-text-align-left\" data-align=\"left\">Objet<\/th><th class=\"has-text-align-left\" data-align=\"left\">Types d&#8217;algorithmes<\/th><\/tr><\/thead><tbody><tr><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"2\">installation pharmaceutique<\/td><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"2\">Risque pour la s\u00e9curit\u00e9<\/td><td class=\"has-text-align-left\" data-align=\"left\">Apprentissage par renforcement (Q-Learning)<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">R\u00e9seau Q profond<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"3\">B\u00e2timent pharmaceutique<\/td><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"3\">Risque pour la sant\u00e9<\/td><td class=\"has-text-align-left\" data-align=\"left\">Sarsa<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">acteur-critique<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">Optimisation de la politique proximale<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"2\">construction intelligente<\/td><td class=\"has-text-align-left\" data-align=\"left\" rowspan=\"2\">Risque pour la s\u00e9curit\u00e9<\/td><td class=\"has-text-align-left\" data-align=\"left\">Gradient de politique d\u00e9terministe profond<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">Apprentissage par renforcement<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><br>Tableau 2.&nbsp;Code&nbsp;de r\u00e9cup\u00e9ration utilis\u00e9 pour la recherche bibliographique.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th class=\"has-text-align-left\" data-align=\"left\">Bases de donn\u00e9es<\/th><th class=\"has-text-align-left\" data-align=\"left\">R\u00e9cup\u00e9rer le code<\/th><\/tr><\/thead><tbody><tr><td class=\"has-text-align-left\" data-align=\"left\">Scopus<\/td><td class=\"has-text-align-left\" data-align=\"left\">TITLE-ABS-KEY (\u00ab Installation pharmaceutique \u00bb OU \u00ab B\u00e2timent pharmaceutique \u00bb OU \u00ab Construction intelligente \u00bb) ET (\u00ab Risque pour la s\u00e9curit\u00e9 \u00bb OU \u00ab Risque pour la sant\u00e9 \u00bb OU \u00ab Risque de s\u00fbret\u00e9 \u00bb) ET (\u00ab Apprentissage par renforcement \u00bb OU \u00ab R\u00e9seau de neurones profond Q \u00bb OU \u00ab Sarsa \u00bb OU \u00ab Apprentissage acteur-critique \u00bb OU \u00ab Optimisation de politique proximale \u00bb OU \u00ab Gradient de politique d\u00e9terministe profond \u00bb OU \u00ab Apprentissage par renforcement \u00bb)<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">Web of Science<\/td><td class=\"has-text-align-left\" data-align=\"left\">TS= (Installation pharmaceutique OU B\u00e2timent pharmaceutique OU Construction intelligente) ET TS= (Risque pour la s\u00e9curit\u00e9 OU Risque pour la sant\u00e9 OU Risque pour la s\u00fbret\u00e9) ET TS= (Apprentissage par renforcement Q OU R\u00e9seau Q profond OU Sarsa OU Apprentissage acteur-critique OU Optimisation de politique proximale OU Gradient de politique d\u00e9terministe profond OU Apprentissage par renforcement)<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"cesectitle0012\">4.1&nbsp;.&nbsp;Analyse de la cooccurrence des mots cl\u00e9s<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Dans cet article, nous effectuons une analyse bibliom\u00e9trique bas\u00e9e sur l&#8217;algorithme de Leiden pour la cooccurrence de mots-cl\u00e9s, \u00e0 l&#8217;aide du logiciel Bibliometrix\u00ae. L&#8217;algorithme de Leiden est un algorithme de d\u00e9couverte de communaut\u00e9s am\u00e9lior\u00e9, con\u00e7u pour optimiser le degr\u00e9 de modularit\u00e9 et pallier certaines lacunes de l&#8217;algorithme de Louvain. Sa formule principale est la suivante&nbsp;:<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" width=\"1024\" height=\"119\" src=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-1024x119.png\" alt=\"\" class=\"wp-image-6661\" style=\"aspect-ratio:8.63770560944749;width:717px;height:auto\" srcset=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-1024x119.png 1024w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-300x35.png 300w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-768x89.png 768w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-20x2.png 20w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-32x4.png 32w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1-600x69.png 600w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.00.29-1.png 1382w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\">Les param\u00e8tres quantitatifs par d\u00e9faut du mod\u00e8le de Leiden sont pr\u00e9sent\u00e9s dans&nbsp;<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#tbl0003\">le Tableau 3.<\/a>&nbsp;Le mod\u00e8le se divise en trois phases principales&nbsp;: d\u00e9placement local des n\u0153uds, raffinement des partitions et agr\u00e9gation des r\u00e9seaux. Les principales \u00e9tapes et formules du mod\u00e8le de Leiden sont les suivantes&nbsp;: (1) D\u00e9placement local des n\u0153uds&nbsp;: une file d\u2019attente contenant tous les n\u0153uds est initialis\u00e9e. Le premier n\u0153ud de la file est ensuite trait\u00e9 et d\u00e9plac\u00e9 vers une nouvelle communaut\u00e9 si ce d\u00e9placement am\u00e9liore le score de la fonction de qualit\u00e9. Si un n\u0153ud est d\u00e9plac\u00e9, ses voisins sont parcourus et ajout\u00e9s \u00e0 la file d\u2019attente aux n\u0153uds n\u2019appartenant pas \u00e0 la nouvelle communaut\u00e9 et qui ne sont pas d\u00e9j\u00e0 pr\u00e9sents. Ces \u00e9tapes sont r\u00e9p\u00e9t\u00e9es jusqu\u2019\u00e0 ce que la file d\u2019attente soit vide. (2) Raffinement des partitions&nbsp;: initialement, chaque n\u0153ud repr\u00e9sente une communaut\u00e9. Les n\u0153uds appartenant \u00e0 une seule communaut\u00e9 sont ensuite fusionn\u00e9s avec d\u2019autres communaut\u00e9s, \u00e0 condition que les deux communaut\u00e9s fusionn\u00e9es le soient. Enfin, une communaut\u00e9 dont le score de qualit\u00e9 est sup\u00e9rieur \u00e0 0 est s\u00e9lectionn\u00e9e al\u00e9atoirement pour la fusion. (3) Agr\u00e9gation du r\u00e9seau&nbsp;: les communaut\u00e9s obtenues lors du raffinement des partitions sont fusionn\u00e9es pour cr\u00e9er un nouveau r\u00e9seau agr\u00e9g\u00e9.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\" id=\"spara011\">Tableau 3.&nbsp;Param\u00e8tres&nbsp;du mod\u00e8le de Leiden.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th class=\"has-text-align-left\" data-align=\"left\">Opacit\u00e9<\/th><th class=\"has-text-align-left\" data-align=\"left\">Taille de l&#8217;\u00e9tiquette<\/th><th class=\"has-text-align-left\" data-align=\"left\">Taille du bord<\/th><th class=\"has-text-align-left\" data-align=\"left\">force de r\u00e9pulsion<\/th><th class=\"has-text-align-left\" data-align=\"left\">Nombre minimal d&#8217;ar\u00eates<\/th><\/tr><\/thead><tbody><tr><td class=\"has-text-align-left\" data-align=\"left\">0,7<\/td><td class=\"has-text-align-left\" data-align=\"left\">3<\/td><td class=\"has-text-align-left\" data-align=\"left\">5<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,1<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Les r\u00e9sultats du regroupement sont pr\u00e9sent\u00e9s dans\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#fig0004\">la figure 4<\/a>\u00a0et\u00a0<a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#tbl0004\">le tableau 4<\/a>\u00a0, et r\u00e9partis en quatre groupes. Le groupe 1 met l&#8217;accent sur le comportement, le mod\u00e8le et le cadre comme \u00e9l\u00e9ments centraux de la recherche appliqu\u00e9e en apprentissage par renforcement. Ce groupe refl\u00e8te les principales applications de l&#8217;apprentissage par renforcement dans la gestion des risques li\u00e9s \u00e0 la construction d&#8217;installations pharmaceutiques, en se concentrant principalement sur la mod\u00e9lisation du comportement du personnel de construction en salle blanche, le respect des proc\u00e9dures d&#8217;installation des \u00e9quipements et la construction d&#8217;un cadre de prise de d\u00e9cision dynamique pour garantir la conformit\u00e9 aux Bonnes Pratiques de Fabrication (BPF). Les mots-cl\u00e9s \u00ab\u00a0comportement\u00a0\u00bb, \u00ab\u00a0mod\u00e8le\u00a0\u00bb et \u00ab\u00a0cadre\u00a0\u00bb pr\u00e9sentent des valeurs \u00e9lev\u00e9es de m\u00e9diation et de PageRank, indiquant que ces concepts sont essentiels pour relier les algorithmes th\u00e9oriques aux besoins sp\u00e9cifiques de l&#8217;industrie pharmaceutique. Les algorithmes d&#8217;apprentissage par renforcement ont obtenu des r\u00e9sultats significatifs en mati\u00e8re de garantie de la qualit\u00e9 de la construction et de pr\u00e9vention des risques de contamination, en capturant l&#8217;impact dynamique du comportement du personnel de construction sur les param\u00e8tres environnementaux des salles blanches et en optimisant la s\u00e9quence des activit\u00e9s cl\u00e9s telles que le soudage des tuyaux et le transfert de la zone aseptique. De plus, la corr\u00e9lation entre \u00ab\u00a0performance\u00a0\u00bb et \u00ab\u00a0algorithme\u00a0\u00bb indique que la recherche se concentre d\u00e9sormais sur l\u2019am\u00e9lioration d\u2019aspects sp\u00e9cifiques de la performance, tels que l\u2019efficacit\u00e9 de la mise en service des lignes de production pharmaceutique et la r\u00e9duction des cycles de validation, plut\u00f4t que sur le d\u00e9veloppement d\u2019algorithmes g\u00e9n\u00e9raux. Les recherches futures devraient viser \u00e0 \u00e9laborer un cadre de mod\u00e9lisation g\u00e9n\u00e9ral int\u00e9grant des facteurs externes comme le \u00ab\u00a0climat\u00a0\u00bb (par exemple, l\u2019impact des fluctuations de temp\u00e9rature et d\u2019humidit\u00e9 ext\u00e9rieures sur les environnements int\u00e9rieurs propres), afin de concevoir des syst\u00e8mes de pr\u00e9diction et de gestion des risques plus robustes et adapt\u00e9s aux installations pharmaceutiques de haute qualit\u00e9.<\/p>\n\n\n<div class=\"wp-block-image\">\n<figure class=\"aligncenter size-large is-resized\"><img decoding=\"async\" width=\"1024\" height=\"517\" src=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-1024x517.png\" alt=\"\" class=\"wp-image-6663\" style=\"aspect-ratio:1.9825750242013553;width:543px;height:auto\" srcset=\"https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-1024x517.png 1024w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-300x151.png 300w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-768x387.png 768w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-20x10.png 20w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-32x16.png 32w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1-600x303.png 600w, https:\/\/sahelib.atatec-design.com\/wp-content\/uploads\/2018\/03\/Capture-decran-2025-12-15-a-16.01.16-1.png 1134w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n<\/div>\n\n\n<p class=\"wp-block-paragraph\" id=\"spara004\">Fig. 4.&nbsp;R\u00e9seau&nbsp;de cooccurrence de mots cl\u00e9s.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\" id=\"spara012\">Tableau 4.&nbsp;R\u00e9sultats&nbsp;de l&#8217;analyse de regroupement.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th class=\"has-text-align-left\" data-align=\"left\">N\u0153ud<\/th><th class=\"has-text-align-left\" data-align=\"left\">Grappe<\/th><th class=\"has-text-align-left\" data-align=\"left\">Interm\u00e9diarit\u00e9<\/th><th class=\"has-text-align-left\" data-align=\"left\">Proximit\u00e9<\/th><th class=\"has-text-align-left\" data-align=\"left\">PageRank<\/th><\/tr><\/thead><tbody><tr><td class=\"has-text-align-left\" data-align=\"left\">comportement<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">24,92429792<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,047619048<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,098891076<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">mod\u00e8le<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">6,928571429<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,038461538<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,085858473<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">cadre<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">9,075702076<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,041666667<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,079130819<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">performance<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">11<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,033333333<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,047488217<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">algorithme<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,032258065<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,035544805<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">climat<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,024390244<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,029006022<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">conception<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">11<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,041666667<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,071218942<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">syst\u00e8me<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">30,76373626<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,052631579<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,123630221<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">s\u00e9curit\u00e9<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">12,30769231<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,041666667<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,086183817<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">optimisation<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,028571429<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,023474778<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">simulation<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,033333333<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,022834954<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">automation<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,035714286<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,037486203<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">r\u00e9seaux<\/td><td class=\"has-text-align-left\" data-align=\"left\">2<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,028571429<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,023957555<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">d\u00e9formation<\/td><td class=\"has-text-align-left\" data-align=\"left\">3<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,058823529<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">excavation<\/td><td class=\"has-text-align-left\" data-align=\"left\">3<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,058823529<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">inspection<\/td><td class=\"has-text-align-left\" data-align=\"left\">4<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,058823529<\/td><\/tr><tr><td class=\"has-text-align-left\" data-align=\"left\">suivi<\/td><td class=\"has-text-align-left\" data-align=\"left\">4<\/td><td class=\"has-text-align-left\" data-align=\"left\">0<\/td><td class=\"has-text-align-left\" data-align=\"left\">1<\/td><td class=\"has-text-align-left\" data-align=\"left\">0,058823529<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"cesectitle0022\">5.&nbsp;Discussions\u200b<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\" id=\"cesectitle0023\">5.1&nbsp;.&nbsp;D\u00e9fis li\u00e9s \u00e0 l&#8217;apprentissage par renforcement<\/h3>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"cesectitle0024\">5.1.1&nbsp;.&nbsp;Difficult\u00e9s li\u00e9es \u00e0 la collecte de donn\u00e9es et \u00e0 la mod\u00e9lisation environnementale<\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">L&#8217;apprentissage par renforcement appliqu\u00e9 aux sc\u00e9narios de construction repose sur une grande quantit\u00e9 de donn\u00e9es \u00e9tat-action-r\u00e9compense. Or, les \u00e9v\u00e9nements \u00e0 haut risque (effondrements, chutes, etc.) sont tr\u00e8s rares en r\u00e9alit\u00e9, ce qui engendre une grave p\u00e9nurie d&#8217;exemples de risques critiques. Par ailleurs, le d\u00e9ploiement de capteurs est limit\u00e9 par des environnements difficiles (poussi\u00e8re, vibrations, etc.), l&#8217;enregistrement manuel est sujet \u00e0 des biais subjectifs et les dispositifs portables peuvent nuire \u00e0 la s\u00e9curit\u00e9 des op\u00e9rations. Plus grave encore, la sp\u00e9cificit\u00e9 de chaque projet rend difficile la migration des donn\u00e9es, obligeant chaque nouveau chantier \u00e0 collecter des donn\u00e9es \u00e0 partir de z\u00e9ro. Ce \u00ab\u00a0d\u00e9marrage \u00e0 froid\u00a0\u00bb limite consid\u00e9rablement l&#8217;utilit\u00e9 de l&#8217;apprentissage par renforcement dans les sc\u00e9narios de construction. Les risques r\u00e9sultent souvent de l&#8217;interaction de facteurs m\u00e9caniques, chimiques, humains et autres facteurs multidomaines (effets combin\u00e9s du vent, des vibrations structurelles et de la manipulation par les travailleurs, par exemple). Les m\u00e9thodes existantes de mod\u00e9lisation des environnements d&#8217;apprentissage par renforcement peinent \u00e0 reproduire fid\u00e8lement ces interactions multiphysiques. Par exemple, les mod\u00e8les simplifi\u00e9s peuvent omettre des m\u00e9canismes de risque cl\u00e9s (comme l\u2019effet des vibrations sur le desserrage des boulons), tandis que les simulations num\u00e9riques haute fid\u00e9lit\u00e9 (comme l\u2019analyse par \u00e9l\u00e9ments finis) sont gourmandes en temps de calcul et ne permettent pas de satisfaire aux interactions en temps r\u00e9el requises pour l\u2019apprentissage par renforcement. Ce paradoxe entre \u00ab\u00a0fid\u00e9lit\u00e9 de mod\u00e9lisation et efficacit\u00e9 de calcul\u00a0\u00bb rend l\u2019environnement d\u2019entra\u00eenement fondamentalement diff\u00e9rent du sc\u00e9nario r\u00e9el.De nombreux chercheurs ont propos\u00e9 la g\u00e9n\u00e9ration de donn\u00e9es synth\u00e9tiques, utilisant la mod\u00e9lisation physique, la simulation proc\u00e9durale ou les mod\u00e8les g\u00e9n\u00e9ratifs (par exemple, les GAN) pour cr\u00e9er des ensembles de donn\u00e9es diversifi\u00e9s et r\u00e9alistes [\u00a0<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bib0152\">152<\/a>\u00a0]. D&#8217;autres ont avanc\u00e9 que l&#8217;apprentissage conjoint, o\u00f9 l&#8217;entra\u00eenement d\u00e9centralis\u00e9 du mod\u00e8le est r\u00e9alis\u00e9 entre diff\u00e9rents chantiers ou entreprises sans \u00e9change des donn\u00e9es originales, est plus efficace, am\u00e9liorant ainsi les capacit\u00e9s de g\u00e9n\u00e9ralisation tout en prot\u00e9geant la confidentialit\u00e9 des donn\u00e9es. En mati\u00e8re de mod\u00e9lisation environnementale, l&#8217;environnement de la construction intelligente est tr\u00e8s dynamique et pr\u00e9sente des interactions physiques complexes, ce qui rend la mod\u00e9lisation pr\u00e9cise extr\u00eamement difficile. De plus, les perturbations externes fr\u00e9quentes (changements m\u00e9t\u00e9orologiques, accidents impr\u00e9vus) rendent la dynamique environnementale difficile \u00e0 pr\u00e9voir, tandis que l&#8217;observabilit\u00e9 partielle (par exemple, l&#8217;\u00e9tat cach\u00e9 des travaux) affaiblit la fiabilit\u00e9 du mod\u00e8le. <\/p>\n\n\n\n<h4 class=\"wp-block-heading\" id=\"cesectitle0025\">5.1.2&nbsp;.&nbsp;Absence de cadre de r\u00e9f\u00e9rence pour les paradigmes d&#8217;application<\/h4>\n\n\n\n<p class=\"wp-block-paragraph\">Actuellement, il n&#8217;existe pas de cadre de r\u00e9f\u00e9rence unifi\u00e9 pour l&#8217;application de l&#8217;apprentissage par renforcement (AR) \u00e0 la gestion des risques dans le secteur de la construction, ce qui engendre un d\u00e9calage important entre la recherche th\u00e9orique et la pratique de l&#8217;ing\u00e9nierie. La communaut\u00e9 acad\u00e9mique privil\u00e9gie les indicateurs de performance des algorithmes (tels que la vitesse de convergence et le rendement), tandis que le domaine de l&#8217;ing\u00e9nierie met davantage l&#8217;accent sur l&#8217;interpr\u00e9tabilit\u00e9, la redondance de s\u00e9curit\u00e9 et la conformit\u00e9 . Ce d\u00e9calage dans les crit\u00e8res d&#8217;\u00e9valuation explique pourquoi de nombreux algorithmes d&#8217;AR, bien que performants dans les publications scientifiques, sont difficiles \u00e0 int\u00e9grer dans les syst\u00e8mes de gestion de l&#8217;ing\u00e9nierie. Plus critique encore, les d\u00e9cisions relatives \u00e0 la s\u00e9curit\u00e9 sur les chantiers n\u00e9cessitent souvent l&#8217;int\u00e9gration des normes sectorielles (par exemple, les normes OSHA) et de l&#8217;expertise, or les cadres d&#8217;AR existants ne disposent pas de m\u00e9canismes permettant d&#8217;int\u00e9grer syst\u00e9matiquement ces connaissances pr\u00e9alables, ce qui conduit \u00e0 des conflits entre les d\u00e9cisions algorithmiques et l&#8217;intuition des ing\u00e9nieurs.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" id=\"cesectitle0031\">6.&nbsp;Conclusion\u200b<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Cet article explore de mani\u00e8re syst\u00e9matique les progr\u00e8s de la recherche, les principaux d\u00e9fis et les strat\u00e9gies d&#8217;am\u00e9lioration de l&#8217;apprentissage par renforcement dans la gestion des risques li\u00e9s \u00e0 la construction d&#8217;installations pharmaceutiques. Les recherches montrent que, malgr\u00e9 les premiers succ\u00e8s obtenus par cette technologie dans la surveillance de la s\u00e9curit\u00e9 des chantiers et la planification des ressources, son application \u00e0 des sc\u00e9narios \u00e0 haut risque, tels que la construction d&#8217;usines pharmaceutiques de haute technologie, de salles blanches et de laboratoires, reste encore \u00e0 ses d\u00e9buts. La recherche actuelle pr\u00e9sente trois limitations majeures&nbsp;: premi\u00e8rement, les exigences sp\u00e9cifiques en mati\u00e8re de contr\u00f4le de la biocontamination, de maintenance dynamique des zones propres et d&#8217;installation aseptique des \u00e9quipements de process n&#8217;ont pas \u00e9t\u00e9 enti\u00e8rement mod\u00e9lis\u00e9es dans les environnements de simulation d&#8217;apprentissage par renforcement&nbsp;; deuxi\u00e8mement, l&#8217;absence d&#8217;un cadre de s\u00e9lection d&#8217;algorithmes pour la v\u00e9rification de la conformit\u00e9 aux BPF et la garantie de la continuit\u00e9 de la production de m\u00e9dicaments rend difficile l&#8217;application des algorithmes courants, tels que DQN et PPO, aux normes de qualit\u00e9 et de s\u00e9curit\u00e9 rigoureuses de la construction pharmaceutique&nbsp;; enfin, le manque de m\u00e9canismes de tra\u00e7abilit\u00e9 et de cadre r\u00e9glementaire industriel pour la prise de d\u00e9cision par apprentissage par renforcement dans le d\u00e9bogage des syst\u00e8mes de process pharmaceutiques et la construction de zones critiques entrave s\u00e9rieusement sa mise en \u0153uvre dans des contextes sensibles, tels que les zones critiques aseptiques.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">R\u00e9f\u00e9rences<\/h2>\n\n\n\n<ol id=\"reference-links-cebibsec1\" class=\"wp-block-list\">\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0001\">[1]<\/a>D&#8217;apr\u00e8s le Quotidien du Peuple, depuis le d\u00e9but du 14e plan quinquennal, l&#8217;industrie pharmaceutique a r\u00e9alis\u00e9 des progr\u00e8s remarquables en mati\u00e8re de d\u00e9veloppement qualitatif, avec un chiffre d&#8217;affaires principal en croissance \u00e0 un taux annuel moyen de 9,3 % (2023).\u00a0<a href=\"https:\/\/www.gov.cn\/yaowen\/liebiao\/202311\/content_6915215.htm\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/www.gov.cn\/yaowen\/liebiao\/202311\/content_6915215.htm<\/a>\u00a0(consult\u00e9 le 29 octobre 2025).<a href=\"https:\/\/scholar.google.com\/scholar?q=People%27s%20Daily%2C%20since%20the%20commencement%20of%20the%2014th%20Five-Year%20Plan%20period%2C%20the%20pharmaceutical%20industry%20has%20achieved%20remarkable%20progress%20in%20high-quality%20development%2C%20with%20main%20business%20revenue%20growing%20at%20an%20average%20annual%20rate%20of%209.3%25.%2C%20(2023).%20https%3A%2F%2Fwww.gov.cn%2Fyaowen%2Fliebiao%2F202311%2Fcontent_6915215.htm%20(accessed%20October%2029%2C%202025).\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0002\">[2]<\/a>B.-G.\u00a0Hwang\u00a0,\u00a0SR\u00a0Thomas\u00a0,\u00a0CH\u00a0CaldasD\u00e9veloppement de mesures de performance pour les projets de construction pharmaceutiqueRevue internationale de gestion de projet\u00a0,\u00a028\u00a0(\u00a02010\u00a0) ,\u00a0p\u00a0.\u00a0265-274\u00a0,\u00a010.1016\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.ijproman.2009.06.004\" target=\"_blank\" rel=\"noreferrer noopener\">\/j.ijproman.2009.06.004<\/a><a href=\"https:\/\/doi.org\/10.1016\/j.ijproman.2009.06.004\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0263786309000751\/pdfft?md5=926c7de39933b509a83db3157eb3f291&amp;pid=1-s2.0-S0263786309000751-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0263786309000751\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-77049091120&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Performance%20metric%20development%20for%20pharmaceutical%20construction%20projects&amp;publication_year=2010&amp;author=B.-G.%20Hwang&amp;author=S.R.%20Thomas&amp;author=C.H.%20Caldas\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0003\">[3]<\/a>Administration d&#8217;\u00c9tat chinoise pour la r\u00e9glementation du march\u00e9 (CSAMR), bonnes pratiques de fabrication des produits pharmaceutiques (r\u00e9vis\u00e9 en 2010), . (2011).\u00a0<a href=\"https:\/\/www.gov.cn\/zhengce\/2021-06\/29\/content_5723551.htm\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/www.gov.cn\/zhengce\/2021-06\/29\/content_5723551.htm<\/a>\u00a0.<a href=\"https:\/\/scholar.google.com\/scholar?q=China%20State%20Administration%20for%20Market%20Regulation%20(CSAMR)%2C%20good%20manufacturing%20practice%20for%20pharmaceutical%20products%20(Revised%202010)%2C%20.%20(2011).%20https%3A%2F%2Fwww.gov.cn%2Fzhengce%2F2021-06%2F29%2Fcontent_5723551.htm.\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0004\">[4]<\/a>J.\u00a0ZhouRecherche sur la m\u00e9thode d&#8217;\u00e9valuation des risques li\u00e9s \u00e0 la gestion de la construction dans les industries pharmaceutiquesJ. Commer. Biotechnol.\u00a0,\u00a027\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a0202-213\u00a0,\u00a010.5912\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.5912\/jcb1112\" target=\"_blank\" rel=\"noreferrer noopener\">jcb1112<\/a><a href=\"https:\/\/doi.org\/10.5912\/jcb1112\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85146461910&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Research%20on%20construction%20management%20risk%20assessment%20method%20of%20Pharmaceutical%20industries&amp;publication_year=2022&amp;author=J.%20Zhou\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0005\">[5]<\/a>S.\u00a0Dabirian\u00a0,\u00a0M.\u00a0Ahmadi\u00a0,\u00a0S.\u00a0AbbaspourAnalyse de l&#8217;impact des politiques financi\u00e8res sur la performance des projets de construction \u00e0 l&#8217;aide de la dynamique des syst\u00e8mesEng., Constr. Architect. Manage..\u00a0,\u00a030\u00a0(\u00a02021\u00a0)\u00a0, pp.\u00a01201\u00a0&#8211;\u00a01221\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1108\/ecam-05-2021-0431\" target=\"_blank\" rel=\"noreferrer noopener\">10.1108\/ecam-05-2021-0431<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Analyzing%20the%20impact%20of%20financial%20policies%20on%20construction%20projects%20performance%20using%20system%20dynamics&amp;publication_year=2021&amp;author=S.%20Dabirian&amp;author=M.%20Ahmadi&amp;author=S.%20Abbaspour\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0006\">[6]<\/a>R.\u00a0EdirisinghePeau num\u00e9rique du chantier\u00a0: technologies de capteurs intelligents pour le chantier intelligent de demainEng., Constr. Architect. Manage..\u00a0,\u00a026\u00a0(\u00a02019\u00a0)\u00a0, pp.\u00a0184\u00a0&#8211;\u00a0223\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1108\/ecam-04-2017-0066\" target=\"_blank\" rel=\"noreferrer noopener\">10.1108\/ecam-04-2017-0066<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0969998819000110\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85054034725&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Digital%20skin%20of%20the%20construction%20site%3A%20smart%20sensor%20technologies%20towards%20the%20future%20smart%20construction%20site&amp;publication_year=2019&amp;author=R.%20Edirisinghe\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0007\">[7]<\/a>M.\u00a0Xu\u00a0,\u00a0X.\u00a0Nie\u00a0,\u00a0H.\u00a0Li\u00a0,\u00a0JCP\u00a0Cheng\u00a0,\u00a0Z.\u00a0MeiChantiers intelligents\u00a0: une approche prometteuse pour am\u00e9liorer la performance de la gestion HSE sur siteJ. Build. Eng.\u00a0,\u00a049\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0104007\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.jobe.2022.104007\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.jobe.2022.104007<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2352710222000201\/pdfft?md5=d66fa9200782f1ccf15f38c3df710be3&amp;pid=1-s2.0-S2352710222000201-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2352710222000201\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85122987476&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Smart%20construction%20sites%3A%20a%20promising%20approach%20to%20improving%20on-site%20HSE%20management%20performance&amp;publication_year=2022&amp;author=M.%20Xu&amp;author=X.%20Nie&amp;author=H.%20Li&amp;author=J.C.P.%20Cheng&amp;author=Z.%20Mei\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0008\">[8]<\/a>V.\u00a0Mnih\u00a0,\u00a0K.\u00a0Kavukcuoglu\u00a0,\u00a0D.\u00a0Silver\u00a0,\u00a0AA\u00a0Rusu\u00a0,\u00a0J.\u00a0Veness\u00a0,\u00a0MG\u00a0Bellemare\u00a0,\u00a0A.\u00a0Graves\u00a0,\u00a0M.\u00a0Riedmiller\u00a0,\u00a0AK\u00a0Fidjeland\u00a0,\u00a0G.\u00a0Ostrovski\u00a0,\u00a0S.\u00a0Petersen\u00a0,\u00a0C.\u00a0Beattie\u00a0,\u00a0A.\u00a0Sadik\u00a0,\u00a0I.\u00a0Antonoglou\u00a0,\u00a0H.\u00a0King\u00a0,\u00a0D.\u00a0Kumaran\u00a0,\u00a0D.\u00a0Wierstra\u00a0,\u00a0S.\u00a0Legg\u00a0,\u00a0D.\u00a0HassabisContr\u00f4le de niveau humain gr\u00e2ce \u00e0 l&#8217;apprentissage par renforcement profondNature\u00a0,\u00a0518\u00a0(\u00a02015\u00a0)\u00a0, p.\u00a0529-533\u00a0,\u00a010.1038\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1038\/nature14236\" target=\"_blank\" rel=\"noreferrer noopener\">nature14236<\/a><a href=\"https:\/\/doi.org\/10.1038\/nature14236\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-84924051598&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Human-level%20control%20through%20deep%20reinforcement%20learning&amp;publication_year=2015&amp;author=V.%20Mnih&amp;author=K.%20Kavukcuoglu&amp;author=D.%20Silver&amp;author=A.A.%20Rusu&amp;author=J.%20Veness&amp;author=M.G.%20Bellemare&amp;author=A.%20Graves&amp;author=M.%20Riedmiller&amp;author=A.K.%20Fidjeland&amp;author=G.%20Ostrovski&amp;author=S.%20Petersen&amp;author=C.%20Beattie&amp;author=A.%20Sadik&amp;author=I.%20Antonoglou&amp;author=H.%20King&amp;author=D.%20Kumaran&amp;author=D.%20Wierstra&amp;author=S.%20Legg&amp;author=D.%20Hassabis\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0009\">[9]<\/a>LP\u00a0Kaelbling\u00a0,\u00a0ML\u00a0Littman\u00a0,\u00a0AW\u00a0MooreApprentissage par renforcement : une \u00e9tudeJ.Artif. Intell. R\u00e9s.\u00a0,\u00a04\u00a0(\u00a01996\u00a0)\u00a0, pp.\u00a0237\u00a0&#8211;\u00a0285\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1613\/jair.301\" target=\"_blank\" rel=\"noreferrer noopener\">10.1613\/jair.301<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-0029679044&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Reinforcement%20Learning%3A%20a%20survey&amp;publication_year=1996&amp;author=L.P.%20Kaelbling&amp;author=M.L.%20Littman&amp;author=A.W.%20Moore\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0010\">[10]<\/a>Y. Han, K. Sun, Z. Zhao, GQ Huang, Planification de l&#8217;am\u00e9nagement modulaire d&#8217;h\u00f4pitaux hors distribution via l&#8217;apprentissage par renforcement profond augment\u00e9 de m\u00e9moire*, dans : 21e Conf\u00e9rence internationale IEEE sur la science et l&#8217;ing\u00e9nierie de l&#8217;automatisation (CASE), IEEE, 2025 : pp. 1292\u20131297.\u00a0<a href=\"https:\/\/doi.org\/10.1109\/case58245.2025.11164092\" target=\"_blank\" rel=\"noreferrer noopener\">10.1109\/case58245.2025.11164092<\/a>\u00a0.<a href=\"https:\/\/scholar.google.com\/scholar?q=Y.%20Han%2C%20K.%20Sun%2C%20Z.%20Zhao%2C%20G.Q.%20Huang%2C%20Out-of-distribution%20modular%20hospital%20fit-Out%20scheduling%20via%20memory-augmented%20Deep%20reinforcement%20learning*%2C%20in%3A%202025%20IEEE%2021st%20International%20Conference%20on%20Automation%20Science%20and%20Engineering%20(CASE)%2C%20IEEE%2C%202025%3A%20pp.%201292%E2%80%931297.%2010.1109%2Fcase58245.2025.11164092.\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0011\">[11]<\/a>JI\u00a0Guerrero\u00a0,\u00a0G.\u00a0Mir\u00f3-Amarante\u00a0,\u00a0A.\u00a0Mart\u00ednSyst\u00e8me d&#8217;aide \u00e0 la d\u00e9cision pour la conception de b\u00e2timents de soins de sant\u00e9 bas\u00e9 sur le raisonnement \u00e0 partir de cas et l&#8217;apprentissage par renforcementExpert. Syst\u00e8me. Appl.\u00a0,\u00a0187\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0116037\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.eswa.2021.116037\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.eswa.2021.116037<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0957417421013828\/pdfft?md5=b40c44306088e9cf92096bb122ed2f94&amp;pid=1-s2.0-S0957417421013828-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0957417421013828\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85116899985&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Decision%20support%20system%20in%20health%20care%20building%20design%20based%20on%20case-based%20reasoning%20and%20reinforcement%20learning&amp;publication_year=2022&amp;author=J.I.%20Guerrero&amp;author=G.%20Mir%C3%B3-Amarante&amp;author=A.%20Mart%C3%ADn\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0012\">[12]<\/a>K. Matsumoto, A. Yamaguchi, T. Oka, M. Yasumoto, S. Hara, M. Lida, M. Teichmann, Approche d&#8217;apprentissage par renforcement bas\u00e9e sur la simulation pour l&#8217;automatisation des machines de construction, dans : ISARC. Actes du Symposium international sur l&#8217;automatisation et la robotique dans la construction, Publications IAARC, Waterloo, 2020 : pp. 457\u2013464.<a href=\"https:\/\/scholar.google.com\/scholar?q=K.%20Matsumoto%2C%20A.%20Yamaguchi%2C%20T.%20Oka%2C%20M.%20Yasumoto%2C%20S.%20Hara%2C%20M.%20Lida%2C%20M.%20Teichmann%2C%20Simulation-based%20reinforcement%20learning%20approach%20towards%20construction%20machine%20automation%2C%20in%3A%20ISARC.%20Proceedings%20of%20the%20International%20Symposium%20on%20Automation%20and%20Robotics%20in%20Construction%2C%20IAARC%20Publications%2C%20Waterloo%2C%202020%3A%20pp.%20457%E2%80%93464.\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0013\">[13]<\/a>K.\u00a0Mason\u00a0,\u00a0S.\u00a0GrijalvaUne revue de l&#8217;apprentissage par renforcement pour la gestion autonome de l&#8217;\u00e9nergie des b\u00e2timentsIng\u00e9nierie \u00e9lectronique informatique\u00a0,\u00a078\u00a0(\u00a02019\u00a0)\u00a0, p.\u00a0300-312\u00a0,\u00a010.1016\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.compeleceng.2019.07.019\" target=\"_blank\" rel=\"noreferrer noopener\">j.compeleceng.2019.07.019<\/a><a href=\"https:\/\/doi.org\/10.1016\/j.compeleceng.2019.07.019\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0045790618333421\/pdfft?md5=4b4df0014e2b508080a6dd254ef0b083&amp;pid=1-s2.0-S0045790618333421-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0045790618333421\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85073644795&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20review%20of%20reinforcement%20learning%20for%20autonomous%20building%20energy%20management&amp;publication_year=2019&amp;author=K.%20Mason&amp;author=S.%20Grijalva\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0014\">[14]<\/a>L.\u00a0Yu\u00a0,\u00a0S.\u00a0Qin\u00a0,\u00a0M.\u00a0Zhang\u00a0,\u00a0C.\u00a0Shen\u00a0,\u00a0T.\u00a0Jiang\u00a0,\u00a0X.\u00a0GuanUne revue de l&#8217;apprentissage par renforcement profond pour la gestion de l&#8217;\u00e9nergie des b\u00e2timents intelligentsIEEE Internet Things Journal\u00a0,\u00a08\u00a0(\u00a02021\u00a0)\u00a0, p.\u00a012046-12063\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/JIOT.2021.3078462\" target=\"_blank\" rel=\"noreferrer noopener\">JIOT.2021.3078462<\/a><a href=\"https:\/\/doi.org\/10.1109\/JIOT.2021.3078462\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85105867665&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20review%20of%20deep%20reinforcement%20learning%20for%20Smart%20building%20energy%20management&amp;publication_year=2021&amp;author=L.%20Yu&amp;author=S.%20Qin&amp;author=M.%20Zhang&amp;author=C.%20Shen&amp;author=T.%20Jiang&amp;author=X.%20Guan\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0015\">[15]<\/a>H.\u00a0Yu\u00a0,\u00a0VWY\u00a0Tam\u00a0,\u00a0X.\u00a0XuUne revue syst\u00e9matique de l&#8217;application de l&#8217;apprentissage par renforcement \u00e0 la simulation du comportement des occupants en lien avec la consommation \u00e9nerg\u00e9tique des b\u00e2timentsEnergy Build.\u00a0,\u00a0312\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0114189\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.enbuild.2024.114189\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.enbuild.2024.114189<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0378778824003050\/pdfft?md5=9d673b50a2772f36930bba8cb7fd4cb3&amp;pid=1-s2.0-S0378778824003050-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0378778824003050\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85191307467&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20systematic%20review%20of%20reinforcement%20learning%20application%20in%20building%20energy-related%20occupant%20behavior%20simulation&amp;publication_year=2024&amp;author=H.%20Yu&amp;author=V.W.Y.%20Tam&amp;author=X.%20Xu\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0016\">[16]<\/a>V.\u00a0Asghari\u00a0,\u00a0Y.\u00a0Wang\u00a0,\u00a0AJ\u00a0Biglari\u00a0,\u00a0S.-C.\u00a0Hsu\u00a0,\u00a0P.\u00a0TangL\u2019apprentissage par renforcement en g\u00e9nie et gestion de la construction : une revueJ. Constr. Eng. Manage\u00a0,\u00a0148\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a003122009\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/(asce)co.1943-7862.0002386\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/(asce)co.1943-7862.0002386<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85138204612&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Reinforcement%20Learning%20in%20Construction%20Engineering%20and%20Management%3A%20a%20review&amp;publication_year=2022&amp;author=V.%20Asghari&amp;author=Y.%20Wang&amp;author=A.J.%20Biglari&amp;author=S.-C.%20Hsu&amp;author=P.%20Tang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0017\">[17]<\/a>R.\u00a0Sarkis-Onofre\u00a0,\u00a0F.\u00a0Catal\u00e1-L\u00f3pez\u00a0,\u00a0E.\u00a0Aromataris\u00a0,\u00a0C.\u00a0LockwoodComment utiliser correctement la d\u00e9claration PRISMASyst. Rev.\u00a0,\u00a010\u00a0(\u00a02021\u00a0)\u00a0, p.\u00a0117\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1186\/s13643-021-01671-z\" target=\"_blank\" rel=\"noreferrer noopener\">10.1186\/s13643-021-01671-z<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85104750068&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=How%20to%20properly%20use%20the%20PRISMA%20statement&amp;publication_year=2021&amp;author=R.%20Sarkis-Onofre&amp;author=F.%20Catal%C3%A1-L%C3%B3pez&amp;author=E.%20Aromataris&amp;author=C.%20Lockwood\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0018\">[18]<\/a>D.\u00a0Moher\u00a0,\u00a0A.\u00a0Liberati\u00a0,\u00a0J.\u00a0Tetzlaff\u00a0,\u00a0DG\u00a0Altman\u00c9l\u00e9ments de rapport privil\u00e9gi\u00e9s pour les revues syst\u00e9matiques et les m\u00e9ta-analyses\u00a0: la d\u00e9claration PRISMAInt. J. Surg.\u00a0,\u00a08\u00a0(\u00a02009\u00a0)\u00a0, p.\u00a0336-341\u00a0,\u00a010.1016\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.ijsu.2010.02.007\" target=\"_blank\" rel=\"noreferrer noopener\">j.ijsu.2010.02.007<\/a><a href=\"https:\/\/doi.org\/10.1016\/j.ijsu.2010.02.007\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Preferred%20reporting%20items%20for%20systematic%20reviews%20and%20meta-analyses%3A%20the%20PRISMA%20statement&amp;publication_year=2009&amp;author=D.%20Moher&amp;author=A.%20Liberati&amp;author=J.%20Tetzlaff&amp;author=D.G.%20Altman\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0019\">[19]<\/a>QJ\u00a0Wen\u00a0,\u00a0ZJ\u00a0Ren\u00a0,\u00a0H.\u00a0Lu\u00a0,\u00a0JF\u00a0WuProgr\u00e8s et tendances de la recherche sur le BIM\u00a0: une analyse de visualisation bas\u00e9e sur la bibliom\u00e9trieAutom. Constr.\u00a0,\u00a0124\u00a0(\u00a02021\u00a0)\u00a0, Article\u00a0103558\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2021.103558\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2021.103558<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521000091\/pdfft?md5=b2bfe1c70f7ff72a50c975f47f1f5184&amp;pid=1-s2.0-S0926580521000091-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521000091\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85099716881&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=The%20progress%20and%20trend%20of%20BIM%20research%3A%20a%20bibliometrics-based%20visualization%20analysis&amp;publication_year=2021&amp;author=Q.J.%20Wen&amp;author=Z.J.%20Ren&amp;author=H.%20Lu&amp;author=J.F.%20Wu\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0020\">[20]<\/a>D.\u00a0Wang\u00a0,\u00a0Y.\u00a0Huangfu\u00a0,\u00a0Z.\u00a0Dong\u00a0,\u00a0Y.\u00a0DongPoints chauds de la recherche et tendances d&#8217;\u00e9volution de la neutralit\u00e9 carbone \u2014 Analyse visuelle des donn\u00e9es bibliom\u00e9triques bas\u00e9e sur CiteSpaceDurabilit\u00e9.\u00a0,\u00a014\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a01078\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/su14031078\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/su14031078<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Research%20hotspots%20and%20evolution%20trends%20of%20Carbon%20neutralityVisual%20analysis%20of%20bibliometrics%20based%20on%20CiteSpace&amp;publication_year=2022&amp;author=D.%20Wang&amp;author=Y.%20Huangfu&amp;author=Z.%20Dong&amp;author=Y.%20Dong\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0021\">[21]<\/a>M.\u00a0Jin\u00a0,\u00a0J.\u00a0Liu\u00a0,\u00a0J.\u00a0Yu\u00a0,\u00a0Q.\u00a0Zhou\u00a0,\u00a0W.\u00a0Wu\u00a0,\u00a0L.\u00a0Fu\u00a0,\u00a0C.\u00a0Yin\u00a0,\u00a0C.\u00a0Fernandez\u00a0,\u00a0H.\u00a0Karimi-Maleh\u00c9volution actuelle et d\u00e9fis futurs des techniques de d\u00e9tection des microplastiques\u00a0: analyse et synth\u00e8se bibliom\u00e9triquesSci. Prog.\u00a0,\u00a0105\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0003685042211321\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1177\/00368504221132151\" target=\"_blank\" rel=\"noreferrer noopener\">10.1177\/00368504221132151<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Current%20development%20and%20future%20challenges%20in%20microplastic%20detection%20techniques%3A%20a%20bibliometrics-based%20analysis%20and%20review&amp;publication_year=2022&amp;author=M.%20Jin&amp;author=J.%20Liu&amp;author=J.%20Yu&amp;author=Q.%20Zhou&amp;author=W.%20Wu&amp;author=L.%20Fu&amp;author=C.%20Yin&amp;author=C.%20Fernandez&amp;author=H.%20Karimi-Maleh\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0022\">[22]<\/a>Y.\u00a0Junjia\u00a0,\u00a0AH\u00a0Alias\u00a0,\u00a0NA\u00a0Haron\u00a0,\u00a0N.Abu\u00a0BakarUne revue syst\u00e9matique bibliom\u00e9trique de l&#8217;\u00e9valuation des risques de s\u00e9curit\u00e9 pour la construction par levage IBSB\u00e2timents\u00a0,\u00a013\u00a0(\u00a02023\u00a0)\u00a0, p.\u00a01853\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/buildings13071853\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/buildings13071853<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85166219048&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20bibliometrics-based%20systematic%20review%20of%20safety%20risk%20assessment%20for%20IBS%20hoisting%20construction&amp;publication_year=2023&amp;author=Y.%20Junjia&amp;author=A.H.%20Alias&amp;author=N.A.%20Haron&amp;author=N.Abu%20Bakar\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0023\">[23]<\/a>Q.\u00a0Xiaoxiang\u00a0,\u00a0Y.\u00a0Junjia\u00a0,\u00a0NA\u00a0Haron\u00a0,\u00a0AH\u00a0Alias\u00a0\u200b\u200b,\u00a0TH\u00a0Law\u00a0,\u00a0N.\u00a0Abu Bakar\u00c9tat des lieux, d\u00e9fis et perspectives d&#8217;avenir de l&#8217;\u00e9valuation des r\u00e9novations de b\u00e2timents \u00e0 \u00e9nergie nette z\u00e9ro\u00a0: une revue syst\u00e9matique bibliom\u00e9triqueEnergies. (B\u00e2le)\u00a0,\u00a017\u00a0(\u00a02024\u00a0)\u00a0, p.\u00a03826\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/en17153826\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/en17153826<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85200748131&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Status%2C%20challenges%20and%20future%20directions%20in%20the%20evaluation%20of%20net-zero%20energy%20building%20retrofits%3A%20a%20bibliometrics-based%20systematic%20review&amp;publication_year=2024&amp;author=Q.%20Xiaoxiang&amp;author=Y.%20Junjia&amp;author=N.A.%20Haron&amp;author=A.H.%20Alias&amp;author=T.H.%20Law&amp;author=N.%20Abu%20Bakar\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0024\">[24]<\/a>Y.\u00a0Junjia\u00a0,\u00a0AH\u00a0Alias\u00a0,\u00a0NA\u00a0Haron\u00a0,\u00a0N.Abu\u00a0BakarApprentissage profond pour la gestion des risques li\u00e9s \u00e0 la s\u00e9curit\u00e9 dans la construction modulaire\u00a0: \u00e9tat des lieux, atouts, d\u00e9fis et perspectives d\u2019avenirAutom. Constr.\u00a0,\u00a0169\u00a0(\u00a02025\u00a0)\u00a0, Article\u00a0105894\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105894\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105894<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524006307\/pdfft?md5=f0876d9b7913d12c9d66de5d9cdac37e&amp;pid=1-s2.0-S0926580524006307-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524006307\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85210275637&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20learning%20for%20safety%20risk%20management%20in%20modular%20construction%3A%20status%2C%20strengths%2C%20challenges%2C%20and%20future%20directions&amp;publication_year=2025&amp;author=Y.%20Junjia&amp;author=A.H.%20Alias&amp;author=N.A.%20Haron&amp;author=N.Abu%20Bakar\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0025\">[25]<\/a>JW\u00a0Drisko\u00a0,\u00a0T.\u00a0MaschiAnalyse de contenuOxford University Press\u00a0,\u00a0New York\u00a0(\u00a02016\u00a0)<a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Content%20Analysis&amp;publication_year=2016&amp;author=J.W.%20Drisko&amp;author=T.%20Maschi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0026\">[26]<\/a>S.\u00a0Golazad\u00a0,\u00a0A.\u00a0Mohammadi\u00a0,\u00a0A.\u00a0Rashidi\u00a0,\u00a0M.\u00a0IlbeigiDu brut au raffin\u00e9\u00a0: pr\u00e9traitement des donn\u00e9es pour les mod\u00e8les d\u2019apprentissage automatique (ML), d\u2019apprentissage profond (DL) et d\u2019apprentissage par renforcement (RL) dans le secteur de la constructionAutom. Constr.\u00a0,\u00a0168\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105844\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105844\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105844<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524005806\/pdfft?md5=a5dab699e7b2969baeb02ac73bd88851&amp;pid=1-s2.0-S0926580524005806-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524005806\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85206910365&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=From%20raw%20to%20refined%3A%20data%20preprocessing%20for%20construction%20machine%20learning%20%2C%20deep%20learning%20%2C%20and%20reinforcement%20learning%20%20models&amp;publication_year=2024&amp;author=S.%20Golazad&amp;author=A.%20Mohammadi&amp;author=A.%20Rashidi&amp;author=M.%20Ilbeigi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0027\">[27]<\/a>B.\u00a0Singh\u00a0,\u00a0R.\u00a0Kumar\u00a0,\u00a0V.P.\u00a0SinghApprentissage par renforcement dans les applications robotiques : une \u00e9tude exhaustiveArtif. Intell. R\u00e9v.\u00a0,\u00a055\u00a0(\u00a02021\u00a0)\u00a0, pp.\u00a0945\u00a0&#8211;\u00a0990\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1007\/s10462-021-09997-9\" target=\"_blank\" rel=\"noreferrer noopener\">10.1007\/s10462-021-09997-9<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Reinforcement%20learning%20in%20robotic%20applications%3A%20a%20comprehensive%20survey&amp;publication_year=2021&amp;author=B.%20Singh&amp;author=R.%20Kumar&amp;author=V.P.%20Singh\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0028\">[28]<\/a>X.\u00a0Lei\u00a0,\u00a0Y.\u00a0Dong\u00a0,\u00a0DM\u00a0Frangopol\u00c9laboration de politiques de maintenance durable du cycle de vie pour les ponts en d\u00e9gradation au niveau du r\u00e9seau \u00e0 l&#8217;aide d&#8217;un auto-encodeur convolutif \u2013 agent d&#8217;apprentissage par renforcement structur\u00e9J. Bridge Ing.\u00a0,\u00a028\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a004023063\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/jbenf2.beeng-6159\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/jbenf2.beeng-6159<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85164247891&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Sustainable%20life-cycle%20maintenance%20policymaking%20for%20network-level%20deteriorating%20bridges%20with%20a%20convolutional%20autoencoderStructured%20reinforcement%20learning%20agent&amp;publication_year=2023&amp;author=X.%20Lei&amp;author=Y.%20Dong&amp;author=D.M.%20Frangopol\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0029\">[29]<\/a>A.\u00a0D\u00e9ria\u00a0,\u00a0P.\u00a0Ghannad\u00a0,\u00a0Y.-C.\u00a0LeeOptimisation dynamique en temps r\u00e9el de l&#8217;affectation d&#8217;unit\u00e9s modulaires \u00e0 des installations hors site dans le cadre de la reconstruction post-catastrophe gr\u00e2ce \u00e0 l&#8217;apprentissage par renforcement profondJ. G\u00e9rer. Ing.\u00a0,\u00a040\u00a0(\u00a02024\u00a0)\u00a0, article\u00a004024021\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/jmenea.meeng-5900\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/jmenea.meeng-5900<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85190967225&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Dynamic%20real-time%20optimization%20of%20modular%20unit%20allocation%20to%20off-site%20facilities%20in%20postdisaster%20reconstruction%20using%20deep%20reinforcement%20learning&amp;publication_year=2024&amp;author=A.%20Deria&amp;author=P.%20Ghannad&amp;author=Y.-C.%20Lee\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0030\">[30]<\/a>GM\u00a0Ali\u00a0,\u00a0A.\u00a0Bouferguene\u00a0,\u00a0M.\u00a0Al-HusseinOptimisation de la disposition des tapis de grue bas\u00e9e sur une approche gloutonne et d&#8217;apprentissage par renforcement \u00e0 base d&#8217;agentsJ. Constr. Eng. Manage\u00a0,\u00a0149\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a004023067\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/jcemd4.coeng-12891\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/jcemd4.coeng-12891<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85161584813&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Crane%20mat%20layout%20optimization%20based%20on%20agent-based%20greedy%20and%20reinforcement-learning%20approach&amp;publication_year=2023&amp;author=G.M.%20Ali&amp;author=A.%20Bouferguene&amp;author=M.%20Al-Hussein\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0031\">[31]<\/a>M.\u00a0Cheng\u00a0,\u00a0DM\u00a0FrangopolUn cadre d\u00e9cisionnel pour la planification de la capacit\u00e9 portante des ponts vieillissants utilisant l&#8217;apprentissage par renforcement profondJ. Comput. Civil Eng.\u00a0,\u00a035\u00a0(\u00a02021\u00a0)\u00a0, Article\u00a004021024\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/(asce)cp.1943-5487.0000991\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/(asce)cp.1943-5487.0000991<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85114093537&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20decision-making%20framework%20for%20load%20rating%20planning%20of%20aging%20bridges%20using%20deep%20reinforcement%20learning&amp;publication_year=2021&amp;author=M.%20Cheng&amp;author=D.M.%20Frangopol\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0032\">[32]<\/a>J.\u00a0Cai\u00a0,\u00a0A.\u00a0Du\u00a0,\u00a0X.\u00a0Liang\u00a0,\u00a0S.\u00a0LiPlanification de trajectoires bas\u00e9e sur la pr\u00e9diction pour une collaboration homme-robot s\u00fbre et efficace dans le secteur de la construction gr\u00e2ce \u00e0 l&#8217;apprentissage par renforcement profondJ. Comput. Civil Eng.\u00a0,\u00a037\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a004022046\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/(asce)cp.1943-5487.0001056\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/(asce)cp.1943-5487.0001056<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85139663489&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Prediction-based%20path%20planning%20for%20safe%20and%20efficient%20HumanRobot%20collaboration%20in%20construction%20via%20deep%20reinforcement%20learning&amp;publication_year=2023&amp;author=J.%20Cai&amp;author=A.%20Du&amp;author=X.%20Liang&amp;author=S.%20Li\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0033\">[33]<\/a>DY\u00a0YangGestion adaptative du cycle de vie des structures \u00e0 grande \u00e9chelle bas\u00e9e sur les risques, utilisant l&#8217;apprentissage par renforcement profond et la mod\u00e9lisation par approximation.J. Ing. M\u00e9canique.\u00a0,\u00a0148\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a004021126\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/(asce)em.1943-7889.0002028\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/(asce)em.1943-7889.0002028<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85118375359&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Adaptive%20risk-based%20life-cycle%20management%20for%20large-scale%20structures%20using%20deep%20reinforcement%20learning%20and%20surrogate%20modeling&amp;publication_year=2022&amp;author=D.Y.%20Yang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0034\">[34]<\/a>SJ\u00a0Hormozabad\u00a0,\u00a0N.\u00a0Jacobs\u00a0,\u00a0MG\u00a0SotoApprentissage par renforcement pour le contr\u00f4le structurel int\u00e9gr\u00e9 et la surveillance de l&#8217;\u00e9tat de sant\u00e9Pract. Period. Struct. Design Constr.\u00a0,\u00a029\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a004024026\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/ppscfx.sceng-1455\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/ppscfx.sceng-1455<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85191344785&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Reinforcement%20learning%20for%20Integrated%20structural%20control%20and%20health%20monitoring&amp;publication_year=2024&amp;author=S.J.%20Hormozabad&amp;author=N.%20Jacobs&amp;author=M.G.%20Soto\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0035\">[35]<\/a>M.\u00a0ElMenshawy\u00a0,\u00a0L.\u00a0Wu\u00a0,\u00a0B.\u00a0Gue\u00a0,\u00a0S.\u00a0AbouRizkAutomatisation de la planification de la fabrication des tron\u00e7ons de tuyauterie pour les projets de construction industrielle modulaire gr\u00e2ce \u00e0 l&#8217;apprentissage par renforcementJ. Comput. Civil Eng.\u00a0,\u00a039\u00a0(\u00a02025\u00a0)\u00a0, Article\u00a004025013\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1061\/jccee5.cpeng-6042\" target=\"_blank\" rel=\"noreferrer noopener\">10.1061\/jccee5.cpeng-6042<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85216212234&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automating%20pipe%20spool%20fabrication%20shop%20scheduling%20for%20modularized%20industrial%20construction%20projects%20using%20reinforcement%20learning&amp;publication_year=2025&amp;author=M.%20ElMenshawy&amp;author=L.%20Wu&amp;author=B.%20Gue&amp;author=S.%20AbouRizk\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0036\">[36]<\/a>W.\u00a0SerranoAlgorithmes d&#8217;apprentissage par renforcement profond dans les infrastructures intelligentesInfrastructures. (B\u00e2le)\u00a0,\u00a04\u00a0(\u00a02019\u00a0)\u00a0, p.\u00a052\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/infrastructures4030052\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/infrastructures4030052<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85078325839&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20reinforcement%20learning%20algorithms%20in%20intelligent%20infrastructure&amp;publication_year=2019&amp;author=W.%20Serrano\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0037\">[37]<\/a>C.\u00a0Zhang\u00a0,\u00a0X.\u00a0Zhou\u00a0,\u00a0C.\u00a0Xu\u00a0,\u00a0Z.\u00a0Wu\u00a0,\u00a0J.\u00a0Liu\u00a0,\u00a0H.\u00a0QiG\u00e9n\u00e9ration automatique de plans de fabrication d&#8217;\u00e9l\u00e9ments pr\u00e9fabriqu\u00e9s en b\u00e9ton \u00e0 partir de donn\u00e9es BIM et d&#8217;un apprentissage par renforcement multi-agentsB\u00e2timents\u00a0,\u00a015\u00a0(\u00a02025\u00a0)\u00a0, p.\u00a0284\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/buildings15020284\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/buildings15020284<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automatic%20generation%20of%20precast%20concrete%20component%20fabrication%20drawings%20based%20on%20BIM%20and%20Multi-Agent%20reinforcement%20learning&amp;publication_year=2025&amp;author=C.%20Zhang&amp;author=X.%20Zhou&amp;author=C.%20Xu&amp;author=Z.%20Wu&amp;author=J.%20Liu&amp;author=H.%20Qi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0038\">[38]<\/a>M.\u00a0Kim\u00a0,\u00a0Y.\u00a0Ham\u00a0,\u00a0C.\u00a0Koo\u00a0,\u00a0T.W.\u00a0KimSimulation des d\u00e9placements des ouvriers sur un chantier de construction par apprentissage par renforcement profond, en tenant compte de leur cognition spatiale et de leur comportement d&#8217;orientation.Autom. Constr.\u00a0,\u00a0147\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0104715\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2022.104715\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2022.104715<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580522005854\/pdfft?md5=0b8552c5a840096eede3cdd46f574655&amp;pid=1-s2.0-S0926580522005854-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580522005854\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85145589638&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Simulating%20travel%20paths%20of%20construction%20site%20workers%20via%20deep%20reinforcement%20learning%20considering%20their%20spatial%20cognition%20and%20wayfinding%20behavior&amp;publication_year=2023&amp;author=M.%20Kim&amp;author=Y.%20Ham&amp;author=C.%20Koo&amp;author=T.W.%20Kim\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0039\">[39]<\/a>P.\u00a0Lin\u00a0,\u00a0L.\u00a0Zhang\u00a0,\u00a0RLK\u00a0Tiong\u00a0,\u00a0X.\u00a0SongApprentissage par renforcement profond guid\u00e9 par la physique pour un fonctionnement am\u00e9lior\u00e9 des tunneliers prenant en compte la stabilit\u00e9 de l&#8217;inertie cin\u00e9tiqueAdv. Eng. Inform.\u00a0,\u00a063\u00a0(\u00a02025\u00a0)\u00a0, Article\u00a0102943\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.aei.2024.102943\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.aei.2024.102943<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034624005949\/pdfft?md5=0cb9b2c2d0a85c46b9795a589192602a&amp;pid=1-s2.0-S1474034624005949-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034624005949\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85209663550&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Physics-guided%20deep%20reinforcement%20learning%20for%20improved%20operation%20of%20TBM%20considering%20kinetic%20inertia%20stability&amp;publication_year=2025&amp;author=P.%20Lin&amp;author=L.%20Zhang&amp;author=R.L.K.%20Tiong&amp;author=X.%20Song\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0040\">[40]<\/a>P.\u00a0Lin\u00a0,\u00a0A.\u00a0Ji\u00a0,\u00a0Y.\u00a0Zhou\u00a0,\u00a0L.\u00a0Zhang\u00a0,\u00a0RLK\u00a0TiongCommande automatis\u00e9e de la position d&#8217;un tunnelier pendant l&#8217;excavation \u00e0 l&#8217;aide d&#8217;un apprentissage par renforcement profond prenant en compte la dynamique spatio-temporelleAppl. Soft. Comput.\u00a0,\u00a0167\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0112234\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.asoc.2024.112234\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.asoc.2024.112234<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1568494624010081\/pdfft?md5=38b802f5d7e86a73449458eb50408100&amp;pid=1-s2.0-S1568494624010081-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1568494624010081\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85204456809&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automated%20position%20control%20of%20tunnel%20boring%20machine%20during%20excavation%20using%20deep%20reinforcement%20learning%20considering%20spatial-temporal%20dynamics&amp;publication_year=2024&amp;author=P.%20Lin&amp;author=A.%20Ji&amp;author=Y.%20Zhou&amp;author=L.%20Zhang&amp;author=R.L.K.%20Tiong\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0041\">[41]<\/a>KM\u00a0Alhamed\u00a0,\u00a0C.\u00a0Iwendi\u00a0,\u00a0AK\u00a0Dutta\u00a0,\u00a0B.\u00a0Almutairi\u00a0,\u00a0H.\u00a0Alsaghier\u00a0,\u00a0S.\u00a0AlmotairiConstruction de b\u00e2timents bas\u00e9e sur la vid\u00e9osurveillance et l&#8217;apprentissage par renforcement profond utilisant un r\u00e9seau \u00e9lectrique intelligentComput. Electric. Eng.\u00a0,\u00a0103\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0108273\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.compeleceng.2022.108273\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.compeleceng.2022.108273<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S004579062200502X\/pdfft?md5=22e879d0186a1d1e35a29df58d0c5151&amp;pid=1-s2.0-S004579062200502X-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S004579062200502X\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85138453139&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Building%20construction%20based%20on%20video%20surveillance%20and%20deep%20reinforcement%20learning%20using%20smart%20grid%20power%20system&amp;publication_year=2022&amp;author=K.M.%20Alhamed&amp;author=C.%20Iwendi&amp;author=A.K.%20Dutta&amp;author=B.%20Almutairi&amp;author=H.%20Alsaghier&amp;author=S.%20Almotairi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0042\">[42]<\/a>M.\u00a0Amani\u00a0,\u00a0R.\u00a0AkhavianOptimisation ergonomique intelligente dans l&#8217;interaction bimanuelle op\u00e9rateur-robot : une approche d&#8217;apprentissage par renforcementAutom. Constr.\u00a0,\u00a0168\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105741\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105741\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105741<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524004771\/pdfft?md5=2b2b2036cf1904a46046f0d067750809&amp;pid=1-s2.0-S0926580524004771-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524004771\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85203868564&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Intelligent%20ergonomic%20optimization%20in%20bimanual%20worker-robot%20interaction%3A%20a%20Reinforcement%20learning%20approach&amp;publication_year=2024&amp;author=M.%20Amani&amp;author=R.%20Akhavian\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0043\">[43]<\/a>D.\u00a0Lee\u00a0,\u00a0S.\u00a0Lee\u00a0,\u00a0N.\u00a0Masoud\u00a0,\u00a0MS\u00a0Krishnan\u00a0,\u00a0VC\u00a0LiApprentissage par renforcement profond pilot\u00e9 par jumeau num\u00e9rique pour l&#8217;allocation adaptative des t\u00e2ches dans la construction robotiqueAdv. Eng. Inform.\u00a0,\u00a053\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0101710\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.aei.2022.101710\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.aei.2022.101710<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034622001689\/pdfft?md5=7b633ac59ee3355dffb75582a7d6202c&amp;pid=1-s2.0-S1474034622001689-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034622001689\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85135829295&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Digital%20twin-driven%20deep%20reinforcement%20learning%20for%20adaptive%20task%20allocation%20in%20robotic%20construction&amp;publication_year=2022&amp;author=D.%20Lee&amp;author=S.%20Lee&amp;author=N.%20Masoud&amp;author=M.S.%20Krishnan&amp;author=V.C.%20Li\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0044\">[44]<\/a>P.\u00a0Lin\u00a0,\u00a0M.\u00a0Wu\u00a0,\u00a0Z.\u00a0Xiao\u00a0,\u00a0RLK\u00a0Tiong\u00a0,\u00a0L.\u00a0ZhangApprentissage par renforcement profond bas\u00e9 sur la physique pour am\u00e9liorer la vitesse d&#8217;avancement et la stabilit\u00e9 des tunneliersAutom. Constr.\u00a0,\u00a0158\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0105234\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2023.105234\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2023.105234<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Physics-informed%20deep%20reinforcement%20learning%20for%20enhancement%20on%20tunnel%20boring%20machines%20advance%20speed%20and%20stability&amp;publication_year=2023&amp;author=P.%20Lin&amp;author=M.%20Wu&amp;author=Z.%20Xiao&amp;author=R.L.K.%20Tiong&amp;author=L.%20Zhang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0045\">[45]<\/a>RK\u00a0Soman\u00a0,\u00a0M.\u00a0Molina-SolanaAutomatisation de la g\u00e9n\u00e9ration de plannings pr\u00e9visionnels pour la construction gr\u00e2ce \u00e0 la v\u00e9rification des contraintes bas\u00e9e sur les donn\u00e9es li\u00e9es et l&#8217;apprentissage par renforcementAutom. Constr.\u00a0,\u00a0134\u00a0(\u00a02022\u00a0)\u00a0, Article\u00a0104069\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2021.104069\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2021.104069<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521005203\/pdfft?md5=e534862bb79633cafcd0b4aec2257143&amp;pid=1-s2.0-S0926580521005203-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521005203\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85120335018&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automating%20look-ahead%20schedule%20generation%20for%20construction%20using%20linked-data%20based%20constraint%20checking%20and%20reinforcement%20learning&amp;publication_year=2022&amp;author=R.K.%20Soman&amp;author=M.%20Molina-Solana\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0046\">[46]<\/a>Y.\u00a0Cao\u00a0,\u00a0W.\u00a0Luo\u00a0,\u00a0Y.\u00a0Xue\u00a0,\u00a0W.\u00a0Lin\u00a0,\u00a0F.\u00a0ZhangCadre d&#8217;apprentissage par renforcement hors ligne bas\u00e9 sur un mod\u00e8le pour l&#8217;optimisation du fonctionnement des tunneliersEspace souterrain\u00a0,\u00a019\u00a0(\u00a02024\u00a0)\u00a0, p.\u00a047-71\u00a0,\u00a010.1016\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.undsp.2024.01.008\" target=\"_blank\" rel=\"noreferrer noopener\">j.undsp.2024.01.008<\/a><a href=\"https:\/\/doi.org\/10.1016\/j.undsp.2024.01.008\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2467967424000540\/pdfft?md5=12e60e379c29962c638a0fecd8e75a42&amp;pid=1-s2.0-S2467967424000540-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2467967424000540\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85196484489&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Model-based%20offline%20reinforcement%20learning%20framework%20for%20optimizing%20tunnel%20boring%20machine%20operation&amp;publication_year=2024&amp;author=Y.%20Cao&amp;author=W.%20Luo&amp;author=Y.%20Xue&amp;author=W.%20Lin&amp;author=F.%20Zhang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0047\">[47]<\/a>D.\u00a0Lee\u00a0,\u00a0M.\u00a0KimSyst\u00e8me de levage de chantier autonome bas\u00e9 sur l&#8217;apprentissage par renforcement profond pour la construction d&#8217;immeubles de grande hauteurAutom. Constr.\u00a0,\u00a0128\u00a0(\u00a02021\u00a0)\u00a0, Article\u00a0103737\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2021.103737\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2021.103737<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521001886\/pdfft?md5=6ed3f0fe696d34eee4d7ae1381acc081&amp;pid=1-s2.0-S0926580521001886-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521001886\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85107116174&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Autonomous%20construction%20hoist%20system%20based%20on%20deep%20reinforcement%20learning%20in%20high-rise%20building%20construction&amp;publication_year=2021&amp;author=D.%20Lee&amp;author=M.%20Kim\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0048\">[48]<\/a>X.\u00a0Liu\u00a0,\u00a0W.\u00a0Zhang\u00a0,\u00a0C.\u00a0Shao\u00a0,\u00a0Y.\u00a0Wang\u00a0,\u00a0Q.\u00a0CongCommande intelligente autonome d&#8217;un tunnelier \u00e0 pression de terre \u00e9quilibr\u00e9e bas\u00e9e sur l&#8217;apprentissage par renforcement profondIng. Appl. Artif. Intell.\u00a0,\u00a0125\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0106702\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.engappai.2023.106702\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.engappai.2023.106702<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0952197623008862\/pdfft?md5=342dc19800d210ce40f519c669719bce&amp;pid=1-s2.0-S0952197623008862-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0952197623008862\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85164666743&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Autonomous%20intelligent%20control%20of%20earth%20pressure%20balance%20shield%20machine%20based%20on%20deep%20reinforcement%20learning&amp;publication_year=2023&amp;author=X.%20Liu&amp;author=W.%20Zhang&amp;author=C.%20Shao&amp;author=Y.%20Wang&amp;author=Q.%20Cong\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0049\">[49]<\/a>Z.\u00a0Zhang\u00a0,\u00a0Z.\u00a0Guo\u00a0,\u00a0H.\u00a0Zheng\u00a0,\u00a0Z.\u00a0Li\u00a0,\u00a0PF\u00a0YuanComposition spatiale architecturale automatis\u00e9e via l&#8217;apprentissage par renforcement profond multi-agents pour la r\u00e9novation de b\u00e2timentsAutom. Constr.\u00a0,\u00a0167\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105702\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105702\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105702<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524004382\/pdfft?md5=a5112bd006589ce11b022b69881c491c&amp;pid=1-s2.0-S0926580524004382-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524004382\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85201445376&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automated%20architectural%20spatial%20composition%20via%20multi-agent%20deep%20reinforcement%20learning%20for%20building%20renovation&amp;publication_year=2024&amp;author=Z.%20Zhang&amp;author=Z.%20Guo&amp;author=H.%20Zheng&amp;author=Z.%20Li&amp;author=P.F.%20Yuan\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0050\">[50]<\/a>Y.\u00a0Pan\u00a0,\u00a0Y.\u00a0Shen\u00a0,\u00a0J.\u00a0Qin\u00a0,\u00a0L.\u00a0ZhangApprentissage par renforcement profond pour l&#8217;optimisation multi-objectif dans la conception de b\u00e2timents \u00e9cologiques bas\u00e9e sur le BIMAutom. Constr.\u00a0,\u00a0166\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105598\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105598\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105598<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003340\/pdfft?md5=aadb5249e67dd70d07f37bead5a13e88&amp;pid=1-s2.0-S0926580524003340-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003340\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85197156715&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20reinforcement%20learning%20for%20multi-objective%20optimization%20in%20BIM-based%20green%20building%20design&amp;publication_year=2024&amp;author=Y.%20Pan&amp;author=Y.%20Shen&amp;author=J.%20Qin&amp;author=L.%20Zhang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0051\">[51]<\/a>CH\u00a0Lin\u00a0,\u00a0B.\u00a0Fu\u00a0,\u00a0L.\u00a0Zhang\u00a0,\u00a0N.\u00a0Li\u00a0,\u00a0GS\u00a0TongConception intelligente de poutres composites acier-b\u00e9ton bas\u00e9e sur l&#8217;apprentissage par renforcement profondStructures\u00a0,\u00a070\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0107666\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.istruc.2024.107666\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.istruc.2024.107666<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2352012424018198\/pdfft?md5=14145991b424cb851b163e769c43a90f&amp;pid=1-s2.0-S2352012424018198-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2352012424018198\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85208145223&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Intelligent%20design%20of%20steelconcrete%20composite%20beams%20based%20on%20deep%20reinforcement%20learning&amp;publication_year=2024&amp;author=C.H.%20Lin&amp;author=B.%20Fu&amp;author=L.%20Zhang&amp;author=N.%20Li&amp;author=G.S.%20Tong\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0052\">[52]<\/a>X.\u00a0Tang\u00a0,\u00a0B.\u00a0Yi\u00a0,\u00a0Z.\u00a0Wang\u00a0,\u00a0MT\u00a0Quasim\u00a0,\u00a0S.\u00a0Basheer\u00a0,\u00a0B.\u00a0LiangOptimisation multi-objectif bas\u00e9e sur l&#8217;apprentissage par renforcement profond pour les b\u00e2timents de soins de sant\u00e9 intelligents en carbone et compatibles avec l&#8217;IIoTIEEE Internet Things Journal\u00a0,\u00a012\u00a0(\u00a02025\u00a0)\u00a0, p.\u00a034768-34779\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/jiot.2025.3579055\" target=\"_blank\" rel=\"noreferrer noopener\">jiot.2025.3579055<\/a><a href=\"https:\/\/doi.org\/10.1109\/jiot.2025.3579055\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-105008393696&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20reinforcement%20learning-based%20Multi-Objective%20optimization%20for%20Carbon%20intelligent%20IIoT-enabled%20healthcare%20buildings&amp;publication_year=2025&amp;author=X.%20Tang&amp;author=B.%20Yi&amp;author=Z.%20Wang&amp;author=M.T.%20Quasim&amp;author=S.%20Basheer&amp;author=B.%20Liang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0053\">[53]<\/a>TF\u00a0Hansen\u00a0,\u00a0GH\u00a0Erharter\u00a0,\u00a0T.\u00a0MarcherVers une politique de changement de fraise TBM pilot\u00e9e par l&#8217;apprentissage par renforcementAutom. Constr.\u00a0,\u00a0165\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105505\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105505\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105505<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524002413\/pdfft?md5=cd54f6ee2a0f24c285a47d2fe4f38908&amp;pid=1-s2.0-S0926580524002413-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524002413\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85195698140&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Towards%20reinforcement%20learning%20-%20driven%20TBM%20cutter%20changing%20policies&amp;publication_year=2024&amp;author=T.F.%20Hansen&amp;author=G.H.%20Erharter&amp;author=T.%20Marcher\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0054\">[54]<\/a>Y.\u00a0Liu\u00a0,\u00a0Y.\u00a0Wang\u00a0,\u00a0Z.\u00a0ZhouSimulation d&#8217;op\u00e9rations coh\u00e9rentes d&#8217;excavatrices lors de travaux de terrassement bas\u00e9e sur l&#8217;apprentissage par renforcementB\u00e2timents\u00a0,\u00a014\u00a0(\u00a02024\u00a0)\u00a0, p.\u00a03270\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/buildings14103270\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/buildings14103270<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85207359881&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Simulation%20of%20coherent%20excavator%20operations%20in%20earthmoving%20tasks%20based%20on%20reinforcement%20learning&amp;publication_year=2024&amp;author=Y.%20Liu&amp;author=Y.%20Wang&amp;author=Z.%20Zhou\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0055\">[55]<\/a>Q.\u00a0Fu\u00a0,\u00a0K.\u00a0Li\u00a0,\u00a0J.\u00a0Chen\u00a0,\u00a0J.\u00a0Wang\u00a0,\u00a0Y.\u00a0Lu\u00a0,\u00a0Y.\u00a0WangPr\u00e9diction de la consommation \u00e9nerg\u00e9tique des b\u00e2timents \u00e0 l&#8217;aide d&#8217;une m\u00e9thode DQN bas\u00e9e sur les for\u00eats profondesB\u00e2timents\u00a0,\u00a012\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a0131\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.3390\/buildings12020131\" target=\"_blank\" rel=\"noreferrer noopener\">10.3390\/buildings12020131<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85123993372&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Building%20energy%20consumption%20prediction%20using%20a%20deep-forest-based%20DQN%20method&amp;publication_year=2022&amp;author=Q.%20Fu&amp;author=K.%20Li&amp;author=J.%20Chen&amp;author=J.%20Wang&amp;author=Y.%20Lu&amp;author=Y.%20Wang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0056\">[56]<\/a>Y.\u00a0Yao\u00a0,\u00a0VWY\u00a0Tam\u00a0,\u00a0J.\u00a0Wang\u00a0,\u00a0KN\u00a0Le\u00a0,\u00a0A.\u00a0ButeraPlanification automatis\u00e9e de la construction utilisant l&#8217;apprentissage par renforcement profond avec \u00e9chantillonnage d&#8217;actions validesAutom. Constr.\u00a0,\u00a0166\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105622\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105622\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105622<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003583\/pdfft?md5=d316d8d4496feb56417c798d9bcfff7a&amp;pid=1-s2.0-S0926580524003583-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003583\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85199755348&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automated%20construction%20scheduling%20using%20deep%20reinforcement%20learning%20with%20valid%20action%20sampling&amp;publication_year=2024&amp;author=Y.%20Yao&amp;author=V.W.Y.%20Tam&amp;author=J.%20Wang&amp;author=K.N.%20Le&amp;author=A.%20Butera\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0057\">[57]<\/a>O.\u00a0Azulay\u00a0,\u00a0A.\u00a0ShapiroContr\u00f4leur de godet de chargeuse sur pneus utilisant l&#8217;apprentissage par renforcement profondIEEE Access\u00a0,\u00a09\u00a0(\u00a02021\u00a0)\u00a0, p.\u00a024145-24154\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/access.2021.3056625\" target=\"_blank\" rel=\"noreferrer noopener\">access.2021.3056625<\/a><a href=\"https:\/\/doi.org\/10.1109\/access.2021.3056625\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85100782018&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Wheel%20loader%20scooping%20controller%20using%20deep%20reinforcement%20learning&amp;publication_year=2021&amp;author=O.%20Azulay&amp;author=A.%20Shapiro\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0058\">[58]<\/a>I.\u00a0Kurinov\u00a0,\u00a0G.\u00a0Orzechowski\u00a0,\u00a0P.\u00a0H\u00e4m\u00e4l\u00e4inen\u00a0,\u00a0A.\u00a0MikkolaExcavatrice automatis\u00e9e bas\u00e9e sur l&#8217;apprentissage par renforcement et la dynamique des syst\u00e8mes multicorpsIEEE Access\u00a0,\u00a08\u00a0(\u00a02020\u00a0)\u00a0, p.\u00a0213998-214006\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/ACCESS.2020.3040246\" target=\"_blank\" rel=\"noreferrer noopener\">ACCESS.2020.3040246<\/a><a href=\"https:\/\/doi.org\/10.1109\/ACCESS.2020.3040246\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85097181516&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automated%20excavator%20based%20on%20reinforcement%20learning%20and%20multibody%20system%20dynamics&amp;publication_year=2020&amp;author=I.%20Kurinov&amp;author=G.%20Orzechowski&amp;author=P.%20H%C3%A4m%C3%A4l%C3%A4inen&amp;author=A.%20Mikkola\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0059\">[59]<\/a>A.\u00a0Ibrahim\u00a0,\u00a0W.\u00a0Torres-Calderon\u00a0,\u00a0M.\u00a0Golparvar-FardApprentissage par renforcement pour la cartographie de la r\u00e9alit\u00e9 de haute qualit\u00e9 des constructions int\u00e9rieures \u00e0 l&#8217;aide de v\u00e9hicules terrestres sans piloteAutom. Constr.\u00a0,\u00a0156\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0105110\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2023.105110\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2023.105110<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580523003709\/pdfft?md5=357539b0eb943aee3ff91ca320d46e5d&amp;pid=1-s2.0-S0926580523003709-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580523003709\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85175185275&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Reinforcement%20learning%20for%20high-quality%20reality%20mapping%20of%20indoor%20construction%20using%20unmanned%20ground%20vehicles&amp;publication_year=2023&amp;author=A.%20Ibrahim&amp;author=W.%20Torres-Calderon&amp;author=M.%20Golparvar-Fard\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0060\">[60]<\/a>T.\u00a0Osa\u00a0,\u00a0M.\u00a0AizawaApprentissage par renforcement profond avec entra\u00eenement antagoniste pour l&#8217;excavation automatis\u00e9e \u00e0 l&#8217;aide d&#8217;images de profondeurIEEE Access\u00a0,\u00a010\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a04523-4535\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/access.2022.3140781\" target=\"_blank\" rel=\"noreferrer noopener\">access.2022.3140781<\/a><a href=\"https:\/\/doi.org\/10.1109\/access.2022.3140781\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85122896475&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20reinforcement%20learning%20with%20adversarial%20training%20for%20automated%20excavation%20using%20depth%20images&amp;publication_year=2022&amp;author=T.%20Osa&amp;author=M.%20Aizawa\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0061\">[61]<\/a>AA\u00a0Apolinarska\u00a0,\u00a0M.\u00a0Pacher\u00a0,\u00a0H.\u00a0Li\u00a0,\u00a0N.\u00a0C\u00f4t\u00e9\u00a0,\u00a0R.\u00a0Pastrana\u00a0,\u00a0F.\u00a0Gramazio\u00a0,\u00a0M.\u00a0KohlerAssemblage robotis\u00e9 de joints de bois par apprentissage par renforcementAutom. Constr.\u00a0,\u00a0125\u00a0(\u00a02021\u00a0)\u00a0, Article\u00a0103569\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2021.103569\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2021.103569<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521000200\/pdfft?md5=57a5d2191810946607da1cfbd1edc2fc&amp;pid=1-s2.0-S0926580521000200-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580521000200\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85101596404&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Robotic%20assembly%20of%20timber%20joints%20using%20reinforcement%20learning&amp;publication_year=2021&amp;author=A.A.%20Apolinarska&amp;author=M.%20Pacher&amp;author=H.%20Li&amp;author=N.%20Cote&amp;author=R.%20Pastrana&amp;author=F.%20Gramazio&amp;author=M.%20Kohler\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0062\">[62]<\/a>C.\u00a0Jiang\u00a0,\u00a0X.\u00a0Li\u00a0,\u00a0J.-R.\u00a0Lin\u00a0,\u00a0M.\u00a0Liu\u00a0,\u00a0Z.\u00a0MaContr\u00f4le adaptatif des flux de ressources pour optimiser les travaux de construction et les flux de tr\u00e9sorerie gr\u00e2ce \u00e0 l&#8217;apprentissage par renforcement profond en ligneAutom. Constr.\u00a0,\u00a0150\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0104817\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2023.104817\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2023.104817<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580523000778\/pdfft?md5=656cbe9951f0d8532e523db428bbfd22&amp;pid=1-s2.0-S0926580523000778-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580523000778\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85149317960&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Adaptive%20control%20of%20resource%20flow%20to%20optimize%20construction%20work%20and%20cash%20flow%20via%20online%20deep%20reinforcement%20learning&amp;publication_year=2023&amp;author=C.%20Jiang&amp;author=X.%20Li&amp;author=J.-R.%20Lin&amp;author=M.%20Liu&amp;author=Z.%20Ma\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0063\">[63]<\/a>AY\u00a0Yasutomi\u00a0,\u00a0H.\u00a0Ichiwara\u00a0,\u00a0H.\u00a0Ito\u00a0,\u00a0H.\u00a0Mori\u00a0,\u00a0T.\u00a0OgataAttention visuo-spatiale et apprentissage par renforcement bas\u00e9 sur les donn\u00e9es proprioceptives pour une t\u00e2che robuste d&#8217;insertion de cheville dans un trou dans des conditions variablesRobot IEEE. Automatique. Lett.\u00a0,\u00a08\u00a0(\u00a02023\u00a0)\u00a0, pp.\u00a01834\u00a0&#8211;\u00a01841\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1109\/lra.2023.3243526\" target=\"_blank\" rel=\"noreferrer noopener\">10.1109\/lra.2023.3243526<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85148458598&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Visual%20spatial%20attention%20and%20proprioceptive%20data-driven%20reinforcement%20learning%20for%20robust%20peg-in-hole%20task%20under%20variable%20conditions&amp;publication_year=2023&amp;author=A.Y.%20Yasutomi&amp;author=H.%20Ichiwara&amp;author=H.%20Ito&amp;author=H.%20Mori&amp;author=T.%20Ogata\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0064\">[64]<\/a>A.\u00a0Zhu\u00a0,\u00a0T.\u00a0Dai\u00a0,\u00a0G.\u00a0Xu\u00a0,\u00a0P.\u00a0Pauwels\u00a0,\u00a0BD\u00a0Vries\u00a0,\u00a0M.\u00a0FangApprentissage par renforcement profond pour la planification d&#8217;assemblage en temps r\u00e9el dans la construction pr\u00e9fabriqu\u00e9e robotis\u00e9eIEEE Trans. Autom. Sci. Eng.\u00a0,\u00a020\u00a0(\u00a02023\u00a0)\u00a0,\u00a0p.\u00a01515-1526\u00a0,\u00a010.1109\u00a0<a href=\"https:\/\/doi.org\/10.1109\/tase.2023.3236805\" target=\"_blank\" rel=\"noreferrer noopener\">\/tase.2023.3236805<\/a><a href=\"https:\/\/doi.org\/10.1109\/tase.2023.3236805\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85147277557&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Deep%20reinforcement%20learning%20for%20real-time%20assembly%20planning%20in%20robot-based%20prefabricated%20construction&amp;publication_year=2023&amp;author=A.%20Zhu&amp;author=T.%20Dai&amp;author=G.%20Xu&amp;author=P.%20Pauwels&amp;author=B.D.%20Vries&amp;author=M.%20Fang\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0065\">[65]<\/a>J.\u00a0Liu\u00a0,\u00a0P.\u00a0Liu\u00a0,\u00a0L.\u00a0Feng\u00a0,\u00a0W.\u00a0Wu\u00a0,\u00a0D.\u00a0Li\u00a0,\u00a0YF\u00a0ChenR\u00e9solution automatis\u00e9e des conflits pour la conception des armatures en acier dans les ossatures en b\u00e9ton via l&#8217;apprentissage par renforcement Q-learning et la mod\u00e9lisation des informations du b\u00e2timentAutom. Constr.\u00a0,\u00a0112\u00a0(\u00a02020\u00a0)\u00a0, Article\u00a0103062\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2019.103062\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2019.103062<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580519304753\/pdfft?md5=02d5647387867895eb88659de8d4c122&amp;pid=1-s2.0-S0926580519304753-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580519304753\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85078657649&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Automated%20clash%20resolution%20for%20reinforcement%20steel%20design%20in%20concrete%20frames%20via%20Q-learning%20and%20Building%20Information%20modeling&amp;publication_year=2020&amp;author=J.%20Liu&amp;author=P.%20Liu&amp;author=L.%20Feng&amp;author=W.%20Wu&amp;author=D.%20Li&amp;author=Y.F.%20Chen\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0066\">[66]<\/a>W.\u00a0Chen\u00a0,\u00a0Z.\u00a0Zhang\u00a0,\u00a0D.\u00a0Tang\u00a0,\u00a0C.\u00a0Liu\u00a0,\u00a0Y.\u00a0Gui\u00a0,\u00a0Q.\u00a0Nie\u00a0,\u00a0Z.\u00a0Zhao\u00c9valuation d&#8217;un algorithme d&#8217;apprentissage par renforcement bas\u00e9 sur LSTM-PPO pour r\u00e9soudre le probl\u00e8me d&#8217;ordonnancement dynamique d&#8217;atelierIng\u00e9nierie industrielle et informatique\u00a0,\u00a0197\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0110633\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.cie.2024.110633\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.cie.2024.110633<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0360835224007551\/pdfft?md5=36b995e9e081140b536763d9c7e4bb3c&amp;pid=1-s2.0-S0360835224007551-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0360835224007551\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85209100572&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Probing%20an%20LSTM-PPO-based%20reinforcement%20learning%20algorithm%20to%20solve%20dynamic%20job%20shop%20scheduling%20problem&amp;publication_year=2024&amp;author=W.%20Chen&amp;author=Z.%20Zhang&amp;author=D.%20Tang&amp;author=C.%20Liu&amp;author=Y.%20Gui&amp;author=Q.%20Nie&amp;author=Z.%20Zhao\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0067\">[67]<\/a>MO\u00a0Ahmed\u00a0,\u00a0IH\u00a0El-adawayMod\u00e9lisation int\u00e9gr\u00e9e par th\u00e9orie des jeux et apprentissage par renforcement pour les appels d&#8217;offres en plusieurs \u00e9tapes dans le secteur de la construction et des infrastructuresConstr. Manage. Econ.\u00a0,\u00a041\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a01-25\u00a0,\u00a010.1080\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1080\/01446193.2022.2124528\" target=\"_blank\" rel=\"noreferrer noopener\">01446193.2022.2124528<\/a><a href=\"https:\/\/doi.org\/10.1080\/01446193.2022.2124528\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=An%20integrated%20game-theoretic%20and%20reinforcement%20learning%20modeling%20for%20multi-stage%20construction%20and%20infrastructure%20bidding&amp;publication_year=2022&amp;author=M.O.%20Ahmed&amp;author=I.H.%20El-adaway\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0068\">[68]<\/a>HS\u00a0Fard\u00a0,\u00a0H.\u00a0Parvin\u00a0,\u00a0M.\u00a0MahmoudiAm\u00e9lioration des pr\u00e9dictions de l&#8217;effet de r\u00e9tr\u00e9cissement dans les tunnels rocheux gr\u00e2ce \u00e0 l&#8217;apprentissage Q d&#8217;ensemble et aux cha\u00eenes de Markov en ligneSci. Rep.\u00a0,\u00a014\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a022885\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1038\/s41598-024-72998-5\" target=\"_blank\" rel=\"noreferrer noopener\">10.1038\/s41598-024-72998-5<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85205527674&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Improving%20predictions%20of%20rock%20tunnel%20squeezing%20with%20ensemble%20Q-learning%20and%20online%20Markov%20chain&amp;publication_year=2024&amp;author=H.S.%20Fard&amp;author=H.%20Parvin&amp;author=M.%20Mahmoudi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0069\">[69]<\/a>F.\u00a0Peng\u00a0,\u00a0H.\u00a0Liu\u00a0,\u00a0L.\u00a0ZhengUne m\u00e9thode d&#8217;apprentissage par renforcement hybride Sarsa pour la pr\u00e9vision de la puissance des batteries robotiquesJ. Cent. South. Univ.\u00a0,\u00a030\u00a0(\u00a02023\u00a0)\u00a0, pp.\u00a03867\u00a0&#8211;\u00a03880\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1007\/s11771-023-5451-0\" target=\"_blank\" rel=\"noreferrer noopener\">10.1007\/s11771-023-5451-0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85180366555&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=A%20Sarsa%20reinforcement%20learning%20hybrid%20ensemble%20method%20for%20robotic%20battery%20power%20forecasting&amp;publication_year=2023&amp;author=F.%20Peng&amp;author=H.%20Liu&amp;author=L.%20Zheng\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0070\">[70]<\/a>C.\u00a0Yu\u00a0,\u00a0G.\u00a0Yan\u00a0,\u00a0K.\u00a0Ruan\u00a0,\u00a0X.\u00a0Liu\u00a0,\u00a0C.\u00a0Yu\u00a0,\u00a0X.\u00a0MiUn r\u00e9seau de portes d&#8217;apprentissage par renforcement convolutionnel ensembliste pour la pr\u00e9vision des PM2.5 dans les stations de m\u00e9tro\u00c9valuation des risques li\u00e9s \u00e0 la recherche stochastique et environnementale\u00a0(\u00a02023\u00a0)\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1007\/s00477-023-02564-4\" target=\"_blank\" rel=\"noreferrer noopener\">10.1007\/s00477-023-02564-4<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=An%20ensemble%20convolutional%20reinforcement%20learning%20gate%20network%20for%20metro%20station%20PM2.5%20forecasting&amp;publication_year=2023&amp;author=C.%20Yu&amp;author=G.%20Yan&amp;author=K.%20Ruan&amp;author=X.%20Liu&amp;author=C.%20Yu&amp;author=X.%20Mi\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0071\">[71]<\/a>R.\u00a0Li\u00a0,\u00a0Z.\u00a0ZouAm\u00e9lioration de l&#8217;apprentissage des robots de construction pour les t\u00e2ches collaboratives et \u00e0 long terme gr\u00e2ce \u00e0 l&#8217;apprentissage par imitation g\u00e9n\u00e9rative adverseAdv. Eng. Inform.\u00a0,\u00a058\u00a0(\u00a02023\u00a0)\u00a0, Article\u00a0102140\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.aei.2023.102140\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.aei.2023.102140<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034623002689\/pdfft?md5=2cec67c1cce8131b7b1f21c8b35f9900&amp;pid=1-s2.0-S1474034623002689-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S1474034623002689\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85170101258&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Enhancing%20construction%20robot%20learning%20for%20collaborative%20and%20long-horizon%20tasks%20using%20generative%20adversarial%20imitation%20learning&amp;publication_year=2023&amp;author=R.%20Li&amp;author=Z.%20Zou\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0072\">[72]<\/a>G.\u00a0Zhu\u00a0,\u00a0Z.\u00a0Shen\u00a0,\u00a0L.\u00a0Liu\u00a0,\u00a0S.\u00a0Zhao\u00a0,\u00a0F.\u00a0Ji\u00a0,\u00a0Z.\u00a0Ju\u00a0,\u00a0J.\u00a0SunM\u00e9thode d&#8217;\u00e9vitement d&#8217;obstacles dynamique pour AUV bas\u00e9e sur un algorithme PPO am\u00e9lior\u00e9IEEE Access\u00a0,\u00a010\u00a0(\u00a02022\u00a0)\u00a0, p.\u00a0121340-121351\u00a0,\u00a010.1109\u00a0\/\u00a0<a href=\"https:\/\/doi.org\/10.1109\/access.2022.3223382\" target=\"_blank\" rel=\"noreferrer noopener\">access.2022.3223382<\/a><a href=\"https:\/\/doi.org\/10.1109\/access.2022.3223382\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85142844416&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=AUV%20dynamic%20obstacle%20avoidance%20method%20based%20on%20improved%20PPO%20algorithm&amp;publication_year=2022&amp;author=G.%20Zhu&amp;author=Z.%20Shen&amp;author=L.%20Liu&amp;author=S.%20Zhao&amp;author=F.%20Ji&amp;author=Z.%20Ju&amp;author=J.%20Sun\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0073\">[73]<\/a>D.\u00a0Yao\u00a0,\u00a0G.\u00a0de S. BorjaAm\u00e9lioration de l&#8217;identification des cyber-risques dans le secteur de la construction gr\u00e2ce aux mod\u00e8les de langageAutom. Constr.\u00a0,\u00a0165\u00a0(\u00a02024\u00a0)\u00a0, Article\u00a0105565\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2024.105565\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2024.105565<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003017\/pdfft?md5=9ead31549c3894ef8c01fbe5cb0f7ab8&amp;pid=1-s2.0-S0926580524003017-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S0926580524003017\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85196671325&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Enhancing%20cyber%20risk%20identification%20in%20the%20construction%20industry%20using%20language%20models&amp;publication_year=2024&amp;author=D.%20Yao&amp;author=G.%20de%20S.%20Borja\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0074\">[74]<\/a>CJ\u00a0Liang\u00a0,\u00a0VR\u00a0Kamat\u00a0,\u00a0CC\u00a0MenassaApprendre aux robots \u00e0 effectuer des t\u00e2ches de construction quasi r\u00e9p\u00e9titives par le biais de d\u00e9monstrations humainesAutom. Constr.\u00a0,\u00a0120\u00a0(\u00a02020\u00a0)\u00a0, Article\u00a0103370\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1016\/j.autcon.2020.103370\" target=\"_blank\" rel=\"noreferrer noopener\">10.1016\/j.autcon.2020.103370<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S092658052030950X\/pdfft?md5=c19ab5f0aa699752cf7af1d19754bd2c&amp;pid=1-s2.0-S092658052030950X-main.pdf\" target=\"_blank\" rel=\"noreferrer noopener\">Voir le PDF<\/a><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S092658052030950X\">Voir l&#8217;article\u00a0<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85088870979&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Teaching%20robots%20to%20perform%20quasi-repetitive%20construction%20tasks%20through%20human%20demonstration&amp;publication_year=2020&amp;author=C.J.%20Liang&amp;author=V.R.%20Kamat&amp;author=C.C.%20Menassa\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0075\">[75]<\/a>W.\u00a0Cai\u00a0,\u00a0L.\u00a0Huang\u00a0,\u00a0Z.\u00a0ZouUne approche int\u00e9gr\u00e9e combinant environnements virtuels et apprentissage par renforcement pour former des robots de construction \u00e0 r\u00e9aliser des t\u00e2ches en situation d&#8217;incertitude.Notes de cours en g\u00e9nie civil\u00a0,\u00a0363\u00a0(\u00a02023\u00a0)\u00a0,\u00a0p.\u00a0259-271\u00a0,\u00a010.1007\u00a0<a href=\"https:\/\/doi.org\/10.1007\/978-3-031-34593-7_17\" target=\"_blank\" rel=\"noreferrer noopener\">\/978-3-031-34593-7_17<\/a><a href=\"https:\/\/doi.org\/10.1007\/978-3-031-34593-7_17\" target=\"_blank\" rel=\"noreferrer noopener\"><\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85171991679&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=An%20integrated%20approach%20combining%20virtual%20environments%20and%20reinforcement%20learning%20to%20train%20construction%20robots%20for%20conducting%20tasks%20under%20uncertainties&amp;publication_year=2023&amp;author=W.%20Cai&amp;author=L.%20Huang&amp;author=Z.%20Zou\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.sciencedirect.com\/science\/article\/pii\/S2666188825010950#bbib0076\">[76]<\/a>J.\u00a0Feng\u00a0,\u00a0Z.\u00a0Ren\u00a0,\u00a0W.\u00a0LiDispatching bas\u00e9 sur les risques des r\u00e9seaux \u00e9lectriques int\u00e9grant la corr\u00e9lation spatio-temporelle et reposant sur un algorithme robuste de type acteur-critique soupleIEEE Trans. Power Syst.\u00a0(\u00a02024\u00a0)\u00a0, pp.\u00a01\u00a0&#8211;\u00a014\u00a0,\u00a0<a href=\"https:\/\/doi.org\/10.1109\/tpwrs.2024.3496936\" target=\"_blank\" rel=\"noreferrer noopener\">10.1109\/tpwrs.2024.3496936<\/a><a href=\"https:\/\/www.scopus.com\/inward\/record.url?eid=2-s2.0-85206880356&amp;partnerID=10&amp;rel=R3.0.0\" target=\"_blank\" rel=\"noreferrer noopener\">Voir dans Scopus\u00a0<\/a><a href=\"https:\/\/scholar.google.com\/scholar_lookup?title=Risk-based%20dispatch%20of%20Power%20systems%20incorporating%20spatiotemporal%20correlation%20based%20on%20the%20robust%20soft%20actor-critic%20algorithm&amp;publication_year=2024&amp;author=J.%20Feng&amp;author=Z.%20Ren&amp;author=W.%20Li\" target=\"_blank\" rel=\"noreferrer noopener\">Google Scholar<\/a><\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>R\u00e9sum\u00e9s La construction intelligente d&#8217;installations pharmaceutiques est confront\u00e9e \u00e0 des risques dynamiques et non lin\u00e9aires, et les m\u00e9thodes de gestion traditionnelles peinent \u00e0 r\u00e9pondre aux exigences \u00e9lev\u00e9es de r\u00e9activit\u00e9 et de conformit\u00e9 en temps r\u00e9el. Or, les recherches existantes sur l&#8217;apprentissage par renforcement (AR) dans ce domaine manquent encore d&#8217;une architecture d&#8217;application syst\u00e9matique et de [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":2323,"menu_order":19,"comment_status":"closed","ping_status":"closed","template":"","doc_tag":[],"class_list":["post-2330","docs","type-docs","status-publish","hentry","no-post-thumbnail"],"acf":[],"_links":{"self":[{"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/docs\/2330","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/docs"}],"about":[{"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/types\/docs"}],"author":[{"embeddable":true,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/comments?post=2330"}],"version-history":[{"count":2,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/docs\/2330\/revisions"}],"predecessor-version":[{"id":6664,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/docs\/2330\/revisions\/6664"}],"up":[{"embeddable":true,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/docs\/2323"}],"wp:attachment":[{"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/media?parent=2330"}],"wp:term":[{"taxonomy":"doc_tag","embeddable":true,"href":"https:\/\/sahelib.atatec-design.com\/index.php\/wp-json\/wp\/v2\/doc_tag?post=2330"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}