Actionable cognitive twins for decision making in manufacturing

Rožanec, Jože M.; Lu, Jinzhi; Rupnik, Jan; Škrjanc, Maja; Mladenić, Dunja; Fortuna, Blaž; Zheng, Xiaochen; Kiritsis, Dimitris

doi:10.1080/00207543.2021.2002967

Cited by 45 publications

(16 citation statements)

References 109 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To fulfill the aforementioned integration and cognitive features, semantic technologies (including ontologies and knowledge graphs) are utilized in [25] as potential solutions for realizing augmented cognitive capabilities. Even more advanced are "actionable cognitive twins", which are enhanced not only with cognitive capabilities by knowledge models, but also provide more advanced insights and decision-making support to users [26]. Despite possible benefits of (actionable) cognitive digital twins, we are not using these paradigms in this article, because we perceive cognitive digital twins rather as a vision without open/standardized realization at this moment.…”

Section: Digital Twins For Production Systemsmentioning

confidence: 99%

Digitalized Automation Engineering of Industry 4.0 Production Systems and Their Tight Cooperation with Digital Twins

Novák

Vyskočil

2022

Processes

View full text Add to dashboard Cite

Smart production systems conforming the Industry 4.0 vision are based on subsystems that are integrated in a way that supports high flexibility and re-configurability. Specific components and devices, such as industrial and mobile robots or transport systems, now pose full-blown systems, and the entire Industry 4.0 production system constitutes a system-of-systems. Testing, fine-tuning, and production planning are important tasks in the entire engineering production system life-cycle. All these steps can be significantly supported and improved by digital twins, which are digitalized replicas of physical systems that are synchronized with the real systems at runtime. However, the design and implementation of digital twins for such integrated, yet partly stand-alone, industrial sub-systems can represent challenging and significantly time-consuming engineering tasks. In this article, the problem of the digital twin design for discrete-event production systems is addressed. The article also proposes to utilize a formal description of production resources and related production operations that the resources can perform. An executable version of such formalization can be automatically derived into a form of a digital twin. Such a derived digital twin can be enhanced with operation duration times that are obtained with process mining methods, leading to more realistic simulations for the entire production system. The proposed solution was successfully tested and validated in the Industry 4.0 Testbed, equipped with four robots and a transport system, which is utilized as a use-case in this article.

show abstract

Section: Digital Twins For Production Systemsmentioning

confidence: 99%

Digitalized Automation Engineering of Industry 4.0 Production Systems and Their Tight Cooperation with Digital Twins

Novák

Vyskočil

2022

Processes

View full text Add to dashboard Cite

show abstract

“…The benefits mentioned above are achieved by enhancing the demand forecasting and optimizing the organization and management of materials, suppliers, and inventory [21,22]. To that end, digital twins of existing processes can be created to simulate, test what-if scenarios, and enhance them without disruption of the physical operations [23,24].…”

Section: Related Workmentioning

confidence: 99%

Enriching Artificial Intelligence Explanations with Knowledge Fragments

et al. 2022

Self Cite

View full text Add to dashboard Cite

Artificial intelligence models are increasingly used in manufacturing to inform decision making. Responsible decision making requires accurate forecasts and an understanding of the models’ behavior. Furthermore, the insights into the models’ rationale can be enriched with domain knowledge. This research builds explanations considering feature rankings for a particular forecast, enriching them with media news entries, datasets’ metadata, and entries from the Google knowledge graph. We compare two approaches (embeddings-based and semantic-based) on a real-world use case regarding demand forecasting. The embeddings-based approach measures the similarity between relevant concepts and retrieved media news entries and datasets’ metadata based on the word movers’ distance between embeddings. The semantic-based approach recourses to wikification and measures the Jaccard distance instead. The semantic-based approach leads to more diverse entries when displaying media events and more precise and diverse results regarding recommended datasets. We conclude that the explanations provided can be further improved with information regarding the purpose of potential actions that can be taken to influence demand and to provide “what-if” analysis capabilities.

show abstract

“…Such a knowledge graph provides ground to assess users' perceived quality of forecast explanations, a yet little researched aspect in XAI [19]. To reuse existing ontology concepts [18,5], we use Basic Formal Ontology (BFO) [16] and Industrial Ontologies Foundry (IOF) specifications [9] and import concepts from [15]. Among new concepts we introduced in our ontology are Feature relevance, Forecast explanation, and Feedback.…”

Section: Ontology and Knowledge Graphmentioning

confidence: 99%

XAI-KG: knowledge graph to support XAI and decision-making in manufacturing

Rožanec¹,

Zajec²,

Kenda³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The increasing adoption of artificial intelligence requires accurate forecasts and means to understand the reasoning of artificial intelligence models behind such a forecast. Explainable Artificial Intelligence (XAI) aims to provide cues for why a model issued a certain prediction. Such cues are of utmost importance to decision-making since they provide insights on the features that influenced most certain forecasts and let the user decide if the forecast can be trusted. Though many techniques were developed to explain black-box models, little research was done on assessing the quality of those explanations and their influence on decision-making. We propose an ontology and knowledge graph to support collecting feedback regarding forecasts, forecast explanations, recommended decision-making options, and user actions. This way, we provide means to improve forecasting models, explanations, and recommendations of decision-making options. We tailor the knowledge graph for the domain of demand forecasting and validate it on real-world data.

show abstract

Actionable cognitive twins for decision making in manufacturing

Cited by 45 publications

References 109 publications

Digitalized Automation Engineering of Industry 4.0 Production Systems and Their Tight Cooperation with Digital Twins

Digitalized Automation Engineering of Industry 4.0 Production Systems and Their Tight Cooperation with Digital Twins

Enriching Artificial Intelligence Explanations with Knowledge Fragments

XAI-KG: knowledge graph to support XAI and decision-making in manufacturing

Contact Info

Product

Resources

About