Cognitive Twin

A cognitive twin is a digital entity mounted with decision-making faculties. Beyond simulating physical objects through digital replicas similar to digital twins, a cognitive twin can self-improve and operate autonomously.
Technology Life Cycle

Technology Life Cycle


Initial phase where new technologies are conceptualized and developed. During this stage, technical viability is explored and initial prototypes may be created.

Technology Readiness Level (TRL)

Technology Readiness Level (TRL)

Field Validation

Validation is conducted in relevant environments, where simulations are carried out as close to realistic circumstances.

Technology Diffusion

Technology Diffusion


First to adopt new technologies. They are willing to take risks and are crucial to the initial testing and development of new applications.

Cognitive Twin

A computer simulation mounted on 3D digital models replicating physical entities with self-improvement faculties. Unlike digital twins, which serve as simulation models of physical objects, cognitive twins actually can make decisions and improve their architecture autonomously.

Cognitive twins are, by definition, digital decision-making agents powered with machine-learning algorithms that can repair themselves from the inside out, providing all kinds of insights. In a factory environment, for instance, a cognitive twin of the whole factory could receive IoT data gathered in different sectors, using this background knowledge to come up with improvements without human supervision.

This self-aware digital replica can simulate entire ecosystems where the cognitive twin itself digitally tests many variables. Accordingly, controllers and programmers can build statistical models that identify mathematical relationships using data and then create simulations based on those equations, allowing the measured traits to play out on the screen. Any AI generative solution works by programming constraints into the algorithms and waiting to resolve an equation that obeys the initial code rules. This way, human biases do not limit the final design or specific solution.

Future Perspectives

Being human, we are often blinded by our evolutionary tools of survival, to the point of not being able to imagine different approaches to problems. We tend to build on old concepts without questioning their fundamentals. By allowing an artificial cognitive agent to be creative on its terms, we are freeing humanity of our shortsightedness. Cognitive twins are not replacing humans in any sense; instead, they sum up forces, creating an overall stronger team with humankind.

Image generated by Envisioning using Midjourney

Digital Twins (DT). A DT is a digital duplication of entities with real-time two-way communication enabled between the physical and cyber spaces. It aims to support integration of IoT for connecting the physical and virtual spaces. In the illustrated case the physical twin is defined as an areo-engine, the virtual entities of areo-engine include CAD models, FEM models etc.
Planes and pumps, buildings and bridges; each has unique story to tell. A Digital Twin help us listen, from my keynote at Hannover Messe.
Digital Twins (DTs) mirror physical assets and can be enriched with software layers that provide different capabilities. In the case of actionable cognitive twins (CTs), algorithms provide behavior (make DTs actionable) and a knowledge graph (KG) adds cognitive capabilities. In this paper we present a new ontology that models a shop-floor DT, capturing background knowledge regarding shop-floor assets and actors, data sources, algorithms (with emphasis on artificial intelligence (AI)) and decision-making opportunities as well as their relations. This ontol-ogy can be used to enhance DTs with cognitive capabilities and instantiated to a KG to provide meaningful context to data and algorithm outcomes , enhancing decision-making suggestions. We describe this through two use cases for an automotive parts manufacturing plant in Europe.
Cognitive Twins (CT) are proposed as Digital Twins (DT) withaugmented semantic capabilities for identifying the dynamics of virtual modelevolution, promoting the understanding of interrelationships between virtualmodels and enhancing the decision-making based on DT. The CT ensures thatassets of Internet of Things (IoT) systems are well-managed and concerns beyondtechnical stakeholders are addressed during IoT system development. In thispaper, a Knowledge Graph (KG) centric framework is proposed to develop CT.Based on the framework, a future tool-chain is proposed to develop the CT forthe initiatives of H2020 project FACTLOG. Based on the comparison betweenDT and CT, we infer the CT is a more comprehensive approach to support IoTbased systems development than DT.

Interested in our research?

Read about our services for help with your foresight needs.