Lexikon

Smart Factory Systems

Abstract:

Manufacturing industry has been facing several challenges including sustainability and performance of the production. These challenges are sourced from numerous factors such as aging workforce, changes in landscape of global manufacturing and slow adaption of smart manufacturing by implementing IT in the manufacturing.

In recent years German and U.S governments have established separate initiatives to accelerate the use of IoT and smart analytics technologies in the manufacturing industries and consequently improve the overall performance, quality and controllability of manufacturing process. Smart factory is the integration of all recent IoT technological advances in computer networks, data integration and analytics to bring transparency to the entire manufacturing factories. In this article we review the most recent logistic decisions for making smart factories from idea to reality and then describe the enabling technologies for smart factories.

From Traditional Factories to Smart Factories:

The manufacturing sector showed a tremendous amount of interest in the new conception introduced in 2013 Hannover Fair in Germany. A futuristic plan developed under the auspices of the German Federal Government’s High-Tech Strategy is outlined to be the framework of the fourth industrial revolution. The first revolution occurred by the end of the 18th century with the mechanization of manufacturing processes. Then towards the start of the next century, electricity was utilized to power mass production of goods based on the division of labor (station-oriented). In the 1970s, the third industrial revolution has been recognized with the use of electronics and information technology (IT) to achieve more automation of manufacturing operations. Based on the initiative, the fourth industrial revolution is the integration of interconnected systems and Internet of Things (IoT) in the manufacturing that is named as Industry 4.0. 

On the other hand, U.S government as another global pioneer in manufacturing industry defined the term Cyber-physical systems (CPS). CPS is a complex engineering system that integrates physical, computation and networking, communication processes. CPS can be illustrated as a physical device, object, equipment that is translated into cyberspace as a virtual model. With networking capabilities, the virtual model can monitor and control its physical aspect, while the physical aspect sends data to update its virtual model. Considering the importance of this topic, cyber-physical systems have been called as national research priority of the United States [1] and European research council [2]. The U.S government has recently established four manufacturing hubs including Additional Manufacturing in Ohio, Low-Power Semiconductor Manufacturing in North Carolina, Digital Manufacturing and Design Innovation (DMDI), and Light Weight Materials in Michigan.  In addition, White House has initiated Smart America Challenges based on the advanced Cyber Physical Systems in 2012.  

Successful integration of Industry 4.0 and cyber-physical systems provides significant benefits for the whole manufacturing industry. These benefits can be summarized in one term so-called: Smart Factory [3]. The adoption of smart factory can be a game-changing event that can transform the interaction of engineered systems just as the internet transformed the way people interact with information. To some extent, we are not only living the physical world, but also in the internet (cyber) space.  For example, Facebook is our cyber life which coexists with our real life. Similar concepts and effects also apply to the manufacturing system in a smart factory. Each physical component and machine will have a twin model in the cyber space composed of data generated from sensor networks and manual inputs. Intelligent data informatics process the data in cyber space so that information about the physical components’ health conditions, performance, and risks are calculated and synchronized in real time. 

As smart factories leverage the web of information from interconnected systems to perform highly efficient, agile and flexible, the overall frame work can be divided into three major sections as is defined by Lee et al [4].  These sections are Components, Machines and Production Systems where each of these items brings different levels of understating and transparency to the factory. Smart machines need to use real-time data from their own components and other machines to gain self-awareness and self-comparison. Self-awareness enables machines to assess their own performance and diagnose possible malfunctioning components. Consequently it can predict and prevent potential failure and risk contributions to the final product. Smart machines can further share their information over the cyber-space to compare their performance and productivity with other similar machines. This self-comparison attribute enables machines to adjust their settings and performance properly by the knowledge they gained from their working history. In this environment, the manufacturing system is also able to schedule customized manufacturing criteria for individual machines based on their performance. Consequently, the production system can configure itself to customize production of every single product based on current status of all involving machines in the manufacturing line to guarantee high quality production with the optimum operation costs.  In such a smart factory, the manufacturer is able to meet the customer specifications in any production rate with supporting last minute changes in the production and other flexibilities that are far from happening in traditional factories. 

Neccessary Technologies for the Smart Factory

The smart factory defines a new approach in multi-scale manufacturing by using the most recent IoT and industrial internet technologies which consists of smart sensors & sensing, computing & predictive analytics, and resilient control technologies. These technologies have to be bonded together to acquire, transfer, interpret and analyze the information and control the manufacturing process as intended. As mentioned in the previous section, fulfilling the requirements of smart factory is possible through cyber-physical systems. Both Industry 4.0 and CPS are in their infancy stages and require more in-depth research to establish their practical usage in different sectors. Smart factory as the symbol for using CPS in manufacturing sector is not exempted from the criteria. At current stage, it is required to define applicable frameworks for establishing CPS in manufacturing industry.

Table 1- The comparison between today’s factories versus Industry 4.0-based smart factories
  Data Source Today's Factory Today's Factory Smart Factory (Industry 4.0-based) Smart Factory (Industry 4.0-based)
    Attributes Technologies Attributes Technologies
Component Sensor Precision Smart Sensors and Fault Detection Self-Aware Self-Predict Degradation Monitoring & Remaining Useful Life Prediction
Machine Controller Producibility & Performance Condition-based Monitoring & Diagnostics Self-Aware Self-Predict Self-Compare Predictive Up Time & Failure Prevention
Production Systems Networked Systems Productivity & OEE Lean Operations: Work and Waste Reduction Self-Configure Self-Maintain Self-Organize Worry-free Productivity with Resilient Control Systems

Recently, 5C architecture has been developed and proposed by the author as the general framework for implementing CPS in manufacturing [5]. The proposed 5-level CPS structure as shown in Fig. 1 provides a step-by-step guideline for developing and deploying a cyber-physical system for smart factory.

5C-Level functions (Figure 2)  can be defined as follows:
 

  • Level 1: Connection, requires acquiring accurate and reliable data from machines and their components. Data source can be from IoT-based machine controller, add-on sensors, quality inspections, maintenance logs, and enterprise management systems such as ERP, MES, and CMM. A seamless and tether-free method for data management and communication, proper selection of sensors, and data streaming are important considerations for this level.  At this level, condition-based monitoring system is normally used to monitor machine status.
  • Level 2: Conversion level is the local machine intelligence, where data are processed and converted to meaningful information (such as machine degradation information). Signal processing, feature extraction and commonly used Prognostics and Health Management (PHM) algorithms (such as self-organizing map, logistics regression, support vector machines, etc.) and predictive analytics are integrated in this level. The outputs of this level include but are not limited to machine health related features, health value, and operation regime flag. The goal for this level is to enable self-aware for component and machine level. 
  • Level 3: Cyber level is where all information confluence and is processed. Peer-to-peer comparison, information sharing, collaborative modeling, and time machine records of machine utilization and health condition history are being analyzed. These analytics provide machines with self-comparison ability, where the performance of a single machine can be compared with and rated among the fleet and on the other hand, similarities between machine performance and previous assets (historical information) can be measured to predict the future behavior of the machinery. Historical data also can be used to correlate the interfacial effects of multiple features.  At this level, cyber physical system approach is normally used to assess machine health in different cycles or regimes and further compare with its peers.
  • Level 4: Cognition level generates a thorough knowledge of the monitored system and provides reasoning information to correlate the effect of different components within the system. Proper organizing and presentation of the acquired knowledge for expert users will support to make proper decisions. Inforgraphic APPs can be used to integrate with machine and user-friendly mobile devices such as smart phone.
  • Level 5: Configuration level is the feedback from cyber space to physical space, where actions are taken as either human-in-the loop or a supervisory control to make machines self-configure, self-adaptive, and self-maintain. This stage acts like a resilience control system to apply corrective and preventive decisions that have been made in cognition level.

An exampled of applying the 5C architecture is illustrated in Fig 3.

Design of Data-less & Information Rich Smart Factory

Another extreme advantage of Cyber-Physical System for smart factory is the capability to manage and present data to different decision makers. The smart connection level makes all the data in digital version, sort the data according to their priority, synchronize the data in same time reference, and organize the data according to their correlations. Hence a connected and paper-less data management environment is built. Moreover, since the data stream is processed in real time, the value of information can be secured with timely actions. The cloud computing and storage capabilities of CPS will allow user to get access to information through mobile devices any time and any places. The info-graphs require minimal data volume to access, and the meanings of the information are only understandable by the users. Hence worries of data security will also be reduced. Users can find useful information in Cyber space at different levels of abstraction, ranging from the condition of a machine component to the overall throughput and quality risk of a manufacturing line. The information retrieval and decision-making process has become much easier due to the effective information abstraction and intuitive representations. Hence users will no longer need to deal with raw data, and resolve the information by themselves. Instead, useful information are mined from data continuously in real-time to create information-rich decision environment, and most of the data are only handled once in the whole processing cycle (Fig. 4).  It is what we called Only Handle Information Once (OHIO) philosophy for a worry-free factory.

Conclusion

Manufacturing has been evolved by innovative technologies and inventions during centuries and gained three major revolutions. These technological advancements acted as catalysts for transforming manufacturing into its current stage. In recent years, the astounding advances in information technology, cloud infrastructure, analytics and even social media have paved the way for the next major transformation in this sector. In Industry 4.0, traditional production facilities are converted into Smart Factories that in turn make Smart Products. Therefore, the new business paradigms are being created for Smart Factories. But the journey towards Smart Factories is expected to be gradual and evolutionary. Many of the basic underlying technologies need to be researched and developed. 

In this position paper, we presented a general technological framework of a smart factory.  In addition, we discussed the advantages of smart factory in terms of interpreting data into information. The full adoption of presented framework will help companies improve their global competitiveness, restore their domestic manufacturing industry, and break ground in new market opportunities.  

References:

[1]    H. Gill, “From vision to reality: cyber-physical systems,” Present. HCSS Natl. Work. New Res. Dir. High Confid. Transp. CPS Automotive, Aviat. Rail, pp. 1–29, 2008.

[2]    “European Commision Research and Innovation - horizon 2020,” European Commision Research and Innovation, 2013. [Online]. Available: ec.europa.eu/programmes/horizon2020/en/. [Accessed: 03-03-2015].

[3]    J. Ö. Erik Sundin, Manufacturing Systems and Technologies for the New Frontier, no. Sfb 627. 2008, pp. 537 – 542.

[4]    J. Lee, “Industry 4.0 in Big Data Environment,” German Harting Magazine, pp. 8–10, 2013.

[5]    J. Lee, B. Bagheri, and H.-A. Kao, “A Cyber Physical Systems Architecture for Industry 4.0-based Manufacturing Systems,” Manuf. Lett., vol. 3, pp. 18–23, Dec. 2014.

Autor und Copyright

Jay Lee
Ohio Eminent Scholar and L. W Scott Alter Chair Professor 
in Advanced Manufacturing
Univ. of Cincinnati
E-Mail 

Founding Director of NSF Industry/University Cooperative Research Center on Intelligent Maintenance Systems (IMS)
Univ. of Cincinnati, Univ. of Michigan, Missouri Univ. of S&T, Univ. of Texas at Austin
www.imscenter.net

 

© Springer-Verlag Berlin Heidelberg 2015