User:Ckayay/sandbox

Intelligent Digital Business Twins

The growing adoption of AI has marked the advent of Business Intelligence (BI) 3.0. Consequently, data democratization capabilities have been elevated along with the introduction of automated analytics and insights. Instead of traditional ways, where human experts define queries and mine databases for answers (a labor-intensive and time-consuming activity), organizations are now seeking to democratize data-based decision making for end-users quickly without dependency on data explorers. These solutions would use automation, self-monitoring, and self-learning analytics to readily detect patterns and/or anomalies in data. That comes with also highly intuitive interfaces providing "Google-like" experiences for the enterprise and the automation on the data and insights preparation side.

While this approach can tell the enterprise what happened, why it happened, and what could happen; it cannot go the next level by industrializing the business-objective based autonomous decision-making in controlled environments.This is the next step in BI maturity that s called BI 4.0. BI 4.0 brings full automation on the end-user side as well through Machine Learning (ML) and Real-time Data Analytics.

The Intelligent Digital Business Twins (iDBT) is developed by Cengiz Kayay (Chief Analytics & Digital Architect) to fulfill the BI 4.0 vision using a novel approach, similar to digital twins. Just as digital twins replicate the physical world into data informed digital models and insights, iDBT realizes business entities instead; these can be Product, Customer, Channel, Seller, Competitor, World, etc..

Description

A revolution is occurring as digital technology becomes embedded in every aspect of our lives; from social media to work. With the emergence of the fourth industrial revolution (Industry 4.0), it's becoming apparent that, to move concurrently with the world, enterprises must deliver increased automation, self-monitoring smart machines, and the ability to analyze and diagnose problems without the need for human intervention.

More and more organizations are employing cognitive technologies in the supply chain. i.e. UPS are replacing daily routes with dynamic ones that adjust in real time depending on weather and traffic conditions.

This makes clear the importance of cognitive technologies and ML to handle data and draw conclusions to optimize business operations. General Electric are using data sensors from jet engines, gas turbines, etc... to capture and store as the data in the form of digital twins; which is a data informed model of a complex system like the jet engine. These models can diagnose faults and predict the need for maintenance to ultimately reduce or eliminate unplanned down times. These twins could possess capabilities such as identifying anomalies, plotting trends in machine performance, and learning efficiencies within a machine to use best practices for other machines.

How about businesses? The business can be modeled just like the Industry 4.0 applications are modeling physical assets to test them for the suitability into the real-world. The digital twins would also continuously help predict faults for the machine . aka preventive maintenance.

Using the same analogy, iDBT is developed that represents the related data and insights created automatically from the real-time transactions data. The iDBT would consume this raw data to readily come up with required insights based on the role-based templates that provides insights to the business concept from different perspectives. The encapsulation of data and related intelligence of business concepts and the ability to respond to external queries based on different roles/context and granularity, would help deliver Intelligent Enterprise strategy and agenda to the next steps. The iDBT can be used to model the business and apply load testing based on business objectives and once deployed to production can monitor the related business performance to predict business faults and take autonomous actions on behalf of the business user as preventive maintenance. The iDBT can be deployed to any environment for several online and batch external consumption scenarios.

Use Cases

Business Operations: Any given business can be modeled using iDBT's to enable augmented analytics and prescriptive analytics quickly in the enterprise. The business decisions could be tested against realistic scenarios and automated.
Business Processes: Any given smart business process can be developed using iDBT's. For e.g. new credit card application process could use the Customer, Payment, Credit Card iDBT's which already delivers the required metrics, analytics and insights to make the required decisions.

Related technologies

References

Saddik, A. El (April 2018). "Digital Twins: The Convergence of Multimedia Technologies". IEEE MultiMedia. 25 (2): 87–92. doi:10.1109/MMUL.2018.023121167. ISSN 1070-986X. S2CID 51922497.
"Minds + Machines: Meet A Digital Twin". Youtube. GE Digital. Retrieved 26 July 2017.
"Introduction to Digital Twin: Simple, but detailed". Youtube. IBM Watson Internet of Things. Retrieved 27 June 2017.
Digital Twins promote preventive maintenance - EEJournal, https://www.eejournal.com/article/digital-twins-promote-predictive-maintenance/