The Smart Factory - Conceptual Models Must Adapt

Author photo: Dick Slansky
ByDick Slansky
Category:
Industry Trends

The way we think about things influences our perception of what's possible.  When considering the use of a set of disruptive technologies such as those in that make up the Industrial Internet of Things, are conventional conceptual models for industrial production a help, or a hindrance?

 The concept of the "lights out factory" has often been portrayed as the culmination of factory automation, where robots, automated production systems, intelligent machines, sensors, and equipment produce products without the intervention of human workers or any manual activity. While the reality of the lights out factory remains in the future for manufacturing at this juncture, remarkable progress has been made in automating the production process.  In industries like automotive, electronics and semiconductor, and food and beverage packaging, automation has evolved from moving production lines that ushered in the era of mass production to complex robotic work cells. These work cells are a marvel of integration and orchestration where robots are integrated with automated conveyance, tooling, and fixtures, and actuators that perform multiple assembly functions and tasks. 

Automation Remains a Key Component of the Smart Factory

Today, manufacturers across many industrial sectors use highly automated production systems that communicate across work cells, production lines, and push real-time production information to supervisory levels, operational dashboards, MES, and other operational and business intelligence applications. Moreover, automation suppliers have developed vertical integration architectures that allow production data to flow upward from connected machines, production lines, and work cells; and automation components such sensors, actuators, and drives. In this vertically integrated automation architecture the data continues its upward flow to control levels (PLCs, PACs, CNCs), and higher graduated levels for manufacturing intelligence and manufacturing operations management (MOM).Dick Slansky MOM.gif

The concept driving a vertical automation architecture was that real-time production data and the actionable information that it would generate would move beyond the supervisory and MES layers to the enterprise business levels of a manufacturer. Moving this production information to enterprise business levels would involve interfacing with SCM, CRM, EAM, and ERP applications, and provide a connection between these business and operational systems. However, the promise of moving real-time production data to business application levels, and the enterprise business systems that it would enable and enhance has not been fully realized.

Some of the reasons behind this are more about shifting of priorities around production operations data to enterprise business integration and interface.  Manufacturers continue to see the value of using real-time production data to measure, monitor, and analyze production processes, and, moreover, continue to optimize the production process. However, rather than rely solely on a vertical architecture based on automation systems, manufacturers also must consider a horizontal architecture based on the product lifecycle of design, build, and operations. It is more the case of these vertical and horizontal axes intersecting at key points in the product, process, and production lifecycle, and providing actionable information that will lead to improving and optimizing the overall production lifecycle process. Additionally, this also represents the merging of the virtually simulated environment with physical production systems truly enabling the concept of the digital factory.

The horizontal architecture of today's design/build/operate lifecycle promotes and enables the concept of the digital factory and enterprise. Enterprise software applications like ERP, SCM, and SRM are integral parts of the lifecycle from design to operations, and are dependent on product (BOM/ERP), process, and production information to power the functions of these enterprise solutions. Along with this connection to enterprise business applications, the product and production lifecycle now includes MES and MOM solutions. These operational applications are integrated with the product design and manufacturing process domains of the product lifecycle. With the emergence of advanced analytics applied to data generated from monitoring and measuring the execution of the production process, manufacturers will be able to fully realize the promise of continuous process improvement. Analytics will be one of the keys to monitoring, controlling, optimizing, and enabling the autonomous factory.       Dick PLM.gif

The Smart Connected Factory Embodies a System of Systems

Clearly, in the case of manufacturing, high value production equipment has been heavily instrumented for some time in a closed, hard-wired factory network environment. Industrial sensors, controllers, and networks have proven to be an expensive investment for manufacturers and upgrades to existing facilities are difficult and often interrupt production. The growth of IoT in the consumer product sector has driven down the cost of sensors, embedded intelligence, and communications interfaces through high volume semiconductor manufacturing. On the other hand, industrial equipment continues to be constrained by a very large installed base of legacy equipment based on industry standards and proprietary communications protocols.

The next generation of smart connected factories will be designed and architected as a system of systems involving production systems, automated work stations, and assembly lines. This is the essence of IIoT and smart manufacturing, and will also address the basic principles of Industrie 4.0, where there is communication between intelligent products, machines, and systems. Much of this communication will be wireless and based on open standards. By connecting machines to machines, people to machines, and machines and people to more expanded system of systems, manufacturers can create intelligent networks and factory systems along the entire value chain. This will lead to factory systems that communicate and control each other autonomously with significantly reduced operator intervention.

The concept of a system of systems is the natural extension of systems engineering and model-based design. The basic idea here is that today's factory and its production systems are most efficient and productive when all of the elements and processes that comprise manufacturing are connected through intelligent networks. This would, of course, start with automated production systems and all of the sensors, controllers, and devices connected to machines and production equipment. Next would be factory visibility, HMI/SCADA, dashboards, and factory intelligence systems, all of which provide factory personnel and managers with real-time actionable information. Mobility technology adds a completely new dimension to factory visibility, where people have access to information beyond the production systems and even the factory. Having access to real-time information anywhere and anytime can significantly shorten the time between when a problem occurs and when it's acted upon.  

Another important dimension is asset performance management (APM) which includes asset management and maintenance. While manufacturers have widely accepted the concept of preventative and condition-based monitoring, many are still in the process of implementing these methods. With the Industrial Internet of Things, lower cost and intelligent sensors, wireless connectivity and more advanced analytics tools are making it less expensive and easier to implement and collect actual performance data and monitor the health of factory equipment.

Both conceptual models, the vertical/hierarchical and the horizontal/lifecycle models, provide a solid basis for beginning to think about the transformational possibilities enabled by the advent of the Industrial Internet of Things and Industrie 4.0.  However, both probably need to be expanded a bit to properly treat the new business models and services that IIoT will create.

Posted in IIoT and I4.0 Viewpoints

Engage with ARC Advisory Group

Representative End User Clients
Representative Automation Clients
Representative Software Clients