Article

Cognitive Manufacturing
Embracing Industry 4.0 Era
June 2020 Artemis Veneti,
Big Data Presales Consultant, Telco Software, Intracom Telecom
Manufacturing & Industry 4.0

The cooperation of advanced manufacturing techniques with information technology and data analytics has already shaped the scenery of the fourth industrial revolution (Industry 4.0). Manufacturing systems and their derivatives have surpassed simple connection requirements and instead, they communicate and exchange data to perform intelligent actions. Here are the four main characteristics of Industry 4.0 (Source: Deloitte):

  • Vertical Networking of highly connected smart systems, able to interact with one another and automatically adjust their performance.
  • Horizontal Integration of all ecosystem entities (business partners, vendors, distributors, customers) across geographical borders.
  • Through-engineering across the entire value chain (end-to-end product lifecycle).
  • Acceleration through exponential technologies that are able to serve mass, low cost & time-to-market applications (machine learning, deep learning, advanced robotics, industrial IoT).
Challenges

Undoubtedly, Industry 4.0 implementation barriers are strong and that may explain the non-massive adoption rate from the manufacturing community worldwide. Challenges that may have played a significant role are described below:

  • Difficulties in Data Integration from multiple, different and geographically distributed data sources.
  • Data Ownership Issues when selecting an external (or even worse a cloud) vendor to host & process company’s information.
  • In-House Expertise Absence in order to support disruptive technologies integration.
  • Lack of Consolidated Strategy to substantially support cross-unit collaboration.

Of course, except for the challenges mentioned above, there is a series of different factors that may explain why some industries have become adaptable & flexible to Industry 4.0 technologies integration, such as the automotive one. Demand for customization and quick delivery, evolving & strict quality & safety standards, but mainly the nature of the industry, clearly spell out the adoption rate variance. A particular industry’s needs may be addressed with less Industry 4.0 technologies; not implying that the industry lacks innovation.

Intracom Telecom in industry 4.0

Intracom Telecom constitutes an established global telecommunication systems & solutions vendor, operating for over 40 years in the telecom market & offering a competitive suite of in-house wireless network, software & ICT solutions. In coordination with Intrarom, its subsidiary in Romania where Intracom Telecom production facilities are located, the Company strategically decided on a series of innovative technology transformations, addressing the challenging industrial revolution.

Leveraging its deep expertise in Big Data Analytics & Artificial Intelligence (AI), Intracom Telecom cross-units joined their forces in order to respond to the following Company's objectives:

  • Process & Quality Improvement:
    Optimizing yield and productivity of manufacturing operations, from design through warranty support.
  • Asset Performance Management:
    Improving reliability and performance of equipment and assets through better visibility, predictability and operations.
  • Resource Optimization:
    Optimizing energy efficiency and facility productivity, while reducing costs.
  • Supply Chain Optimization:
    Improving visibility and insights to build a dynamic supply chain that accelerates innovation.
Illustrating Status-Quo

In order to successfully redesign existing processes & introduce new meaningful technologies, it is critical to understand the so called “As-Is” status. Identifying the ecosystem's actors including business objects, roles, activities & interactions, as well as the product life cycle stages, was the building stone of this initiative. At that particular stage, use case & class diagrams, helped the teams to detect the system’s classes, their attributes & interconnections.

Data – Driven Approach

After acquiring the necessary system knowledge accompanied by the valuable feedback from the business & IT stakeholders (regarding existing weaknesses in their day-to-day routine), a data-driven approach was followed.

Different data sources & legacy systems, geographically distributed in both Greece and Romania, hosting a massive amount of semi-structured and unstructured data, had to be thoroughly analyzed, in order to conclude to the new suitable technology infrastructure.

Big Data Analytics & AI
  1. a staging data infrastructure that could support real-time & easy search and analysis of raw data (ELK Stack), and
  2. a final data infrastructure (big data cluster), able to support complex analytics, BI & AI use cases (Cloudera CDH).

More specifically, the staging data area serves the business owners' need for ad-hoc and instant search of production & quality log files (e.g. in case of a test failure and its root cause), whereas the final data area hosts transformed data structures for advanced analytics, flexible reporting and implementation of competitive AI use cases.

Results & Next Steps

Results so far have shown that business stakeholders have a significantly faster access to the information they need, in the way they need it (personalized, intuitive dashboards). Moreover, product quality checks are now quicker, more precise and with less human intervention or errors. In the meantime, a list of meaningful AI use cases is being designed, including fault prediction, predictive maintenance, generative design and more.

Finally, it is for sure that there is a long road to walk through until we achieve the optimal results; however, we are already witnessing the glaring technology revolution benefits.

Manufacturing 5.0

Industry 4.0 undoubtedly reshaped -and still does- the manufacturing ecosystem. By leveraging modern technologies as the IoT, AI, Big Data, Cloud Computing and Robotic Process Automation, manufacturing firms optimize their internal systems and processes, achieving faster, easier, more valuable and cost-efficient results.

Most Industry 4.0 real-life applications basically aim at product development, inventory management, production & quality amelioration via generative design, demand forecast, image recognition for defect inspection, and predictive maintenance.

Industry 5.0, yet a penetrating trend to the mass market, takes advantage of the previous generation (which is still there), in order to introduce the importance of a closer synergy between human and machine, via collaborative robots. While Industry 4.0 prioritizes optimum quantity and mass production, the fifth industrial revolution leans on life standard, creativity and high-quality customized products. In this way, customers can get value-added, but also human-developed products on-demand, whereas workforce can be occupied to more meaningful tasks.

Intracom Telecom embraces first the ongoing, challenging industry revolutions, by continuously investing in a series of disruptive technology transformations. Our vision is reflected by a Human-Centric, Smart Society solution, based on the intelligent partnership between humans and machines.