Figure 5 - uploaded by Benjamin Knoke
Content may be subject to copyright.
Classification Model of IPS The Classification Model of IPS is comprised of the Level of Intelligence, the addressee of relevant information, and the location of intelligence. It has been summarized in Figure 5.
Source publication
![Figure 2. Classification model of Intelligent Products [13]](profile/Benjamin-Knoke/publication/258884817/figure/fig2/AS:667066239549453@1536052393801/Classification-model-of-Intelligent-Products-13_Q320.jpg)
Along its lifecycle, a product passes through several states, which can be described by their characteristics. This concept of product state characteristics can be utilized in manufacturing process chains to improve the process quality by e.g., increasing transparency. The basic approach of describing a product state by its state-characteristics ca...
Similar publications
A transparent clay manufactured using Carbopol® UltrezTM 10 (simplified as U10) was introduced, and its manufacturing processes were briefly described. Both relative transparency (RT) and modulation transfer function (MTF) methods were used to quantify the optical character variation via soil thickness. The transparency of this new transparent clay...
Citations
... Dans (Knoke et al. 2013), les auteurs explorent la possibilité pour les PI de mesurer, stocker , traiter et communiquer les informations relatives à une système de production dans un environnement manufacturier collaboratif. ...
Ces dernières années ont vue l'apparition de nouvelles approches de pilotage des systèmes de production. Dans cet article, nous présentons une approche offrant un contrôle distribué du système de pilotage. Grace au concept de " produit intelligent " il est possible d'éviter une centralisation des décisions. Une des originalités de l'approche présentée est l'utilisation conjointe des techniques de produit intelligent avec les méthodes d'apprentissage machine. Ce mécanisme est utilisé pour déterminer la meilleure stratégie d'ordonnancement pour assurer une performance globale du système de production. Mots clés—Produit intelligent, Ordonnancement, Q-learning, Multi-Agent, Atelier flexible.
... The stigmergic routing approach uses a technique of reinforcement learning, which allows finding the best path by exploiting the routing experiences of the IPs crossing the FMS. In [15], the authors explore which existent IPs support the measuring, storage, and communication of knowledge based on state-characteristics and process parameters within a collaborative manufacturing environment. ...
... -LE MOIGNE [13] distinguishes seven levels of intelligence, from a completely passive object (such as a raw material) to self-completing objects. The objects in the upper levels exhibit cognitive capabilities, although the concept of self-finalized objects seems a bit futuristic. ...
... The stigmergic routing approach uses a reinforcement learning technique, which allows the best path to be found by exploiting the routing experiences of the IPs going through the FMS. In [13], the authors explore to what extent IPs support the measurement, storage, and communication of knowledge based on state characteristics and process parameters within a collaborative manufacturing environment. ...
The needs of flexibility, agility and adaptation capabilities for modern manufacturing systems increase constantly. In this paper, we propose an original approach combining active/intelligent product architecture with learning mechanism to assure flexibility and agility to the overall manufacturing system. Using learning approaches as Reinforcement Learning (RL) mechanism, an active product can be able to reuse learned experiences to enhance its decisional performances. A contextualization method is proposed to improve the decision making of the product for scheduling tasks. The approach is then applied to a case study using a multi-agent simulation platform.
... Dans (Knoke et al. 2013), les auteurs explorent la possibilité pour les PI de mesurer, stocker , traiter et communiquer les informations relatives à une système de production dans un environnement manufacturier collaboratif. ...
Ces dernières années ont vue l’apparition de nouvelles approches de pilotage des systèmes de production. Dans cet article, nous présentons une approche offrant un contrôle distribué du système de pilotage. Grace au concept de “produit intelligent” il est possible d’éviter une centralisation des décisions. La centralisation affecte fréquemment de façon négative la réactivité du système de contrôle. Nous distinguons alors deux niveaux systémiques ; l’un est une microstructure individuelle qui décrit la composition interne de l’entité intelligente ; la seconde est un modèle collectif appelé macrostructure. Cette modélisation des entités individuelles et de leurs relations est purement « hétérarchique ». Le comportement global est ainsi obtenu par émergence.
