Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Efficient Practices of Cognitive Technology Application for Smart Manufacturing

Efficient Practices of Cognitive Technology Application for Smart Manufacturing Cognitive manufacturing (CM) provides for the merging of sensor-based information, advanced analytics, and cognitive technologies, mainly machine learning in the context of Industry 4.0. Manufacturers apply cognitive technologies to review current business metrics, solve essential business problems, generate new value in their manufacturing data and improve quality. The article investigates four powerful applications for cognitive manu- facturing and their influence on a company`s maintenance. The study aims to observe kinds of cognitive technol- ogy applications for smart manufacturing, distinguish their prospective gains for manufacturers and provide suc- cessful examples of their adoption. The analysis is based on the literature and report review. Assessment of the cases of technology adoption proves that cognitive manufacturing provides both enhanced knowledge manage- ment and helps organizations improve fundamental business measurements, such as productivity, product relia- bility, quality, safety, and yield while reducing downtime and lowering costs. Key words: cognitive manufacturing, cognitive technology, Industry 4.0 INTRODUCTION adopted AI use cases within manufacturing [4, 5]. Across Innovative scientific tendencies, the emergence of manu- manufacturing respondents, they report higher levels of facturing 4.0 and cognitive technologies impact manufac- cost decreases from AI adoption in more than one-and- turing procedures and effectiveness. The fundamental one-half times in the fiscal year 2020 compared to the fis- idea of Industry 4.0 is to adopt the Internet of Things (IoT), cal year 2019 with a dramatic increase of percentage rep- artificial intelligence (AI), and cyber-physical systems thus resenting a decrease of more than 20 per cent while rev- manufacturing process and business process are ex- enue increase remains almost at the same level [4]. tremely integrated to provide flexible and efficient pro- Considering the cognitive technology to be a part of the duction [1, 2]. Manufacturing continues to change as or- larger area of AI and a dramatic increase in technologies ganizations equip their factory with technologies and ap- adoption, cognitive technology application for smart man- ply complex data analytics to improve performance, prod- ufacturing is worth examining. Hence the article aims to uct quality, and the way perception of manufacturing in- investigate four applications for cognitive manufacturing, formation is found and responded to. what areas of manufacturing they can refer to, what are According to the 2022 manufacturing industry outlook by advantages for manufacturers for embedding CT into pro- Deloitte, an increase in investment in artificial intelligence duction and give examples confirming the success of cog- technologies is expected at a compound annual growth nitive technology application. rate above 20% through 2025 [3]. A global transformation LITERATURE REVIEW is in progress to equip manufacturing with AI. Discrete The use of technologies in different fields under Industry manufacturing ranks among the three industries with the 4.0, such as robotics and artificial intelligence, poses a highest levels of contribution in AI including quality man- new requirement [6]. The approach of cognitive manufac- agement and automated preventive maintenance use turing is consonant with such concepts as smart manufac- cases [3]. turing [7], and Industry 4.0 [3, 8, 9]. The authors [8, 9] fig- In the McKinsey Global Survey "The State of AI in 2021" ure out Industry 4.0 as the most essential part of the predictive maintenance and yield, energy, and/or Fourth Industrial Revolution with smart manufacturing throughput optimization are defined to be the most © 2022 Author(s). This is an open access article licensed under the Creative Commons BY 4.0 (https://creativecommons.org/licenses/by/4.0/ 188 Management Systems in Production Engineering 2022, Volume 30, Issue 2 being part of it, overlapping concepts, namely the indus- IoT, with an increase in process efficiency, lower opera- trial internet, the industrial internet of things and intelli- tional costs, increased safety and sustainability, and in- gent manufacturing. creased product quality [19]. Manufacturers adopt cogni- Two major technologies that are pushing Industry 4.0 are tive technologies to detail manufacturing processes and IoT and analytics [10, 11]. IoT allows factories to improve business environments aiming at obtaining information their instrumented and interconnected status with the suited for further manufacturing processes digitizing and ability to provide gained relevant data together from the optimization. Targets for cost-effectiveness, quality, prod- manufacturing environment in real-time. Moens and oth- uct reliability, cost, and time efficiency force organizations ers [12] provide an Industrial Internet of Things (IIoT) sys- to search for ways how to improve their manufacturing tem that includes smart maintenance solutions. They re- processes. veal the robustness and scaling of solutions as well as the The concept of cognitive manufacturing is relevant to the availability of well-trained machine learning models in ability to deal with complex modern production systems case of defect recognition. Industrial IoT establishes new and have a quick reaction to unforeseeable issues in pro- features in industrial automation systems which are de- duction, planning and control systems. Thus, the Cogni- termined by improving system robustness and flexibility tive Factory is assumed to be flexible, adaptable, reliable, [13]. and efficient in various momentary situations [20]. Analytics stands for determining templates in the data, a METHODOLOGY pattern of equipment approach and a forecast of possible The methodology of this article was based on a three-step breakdowns. It implies a variety of statistical techniques approach. The first step included preliminary research on including pattern recognition, text analytics, cluster anal- the concept of cognitive manufacturing and obtaining a ysis, factor analysis, multivariate modelling, multiple re- deeper knowledge of the research area. The second stage gression, forecasting, machine learning, simulation, and consisted of the examination of academic literature and neural networks [14]. The research of Rousopoulou V. and report review on the applications for cognitive manufac- others introduce cognitive analytics, self-and autono- turing leading to a restraint of the scope of a theoretical mous-learned system bearing predictive maintenance so- foundation. The last one covered case studies of cognitive lutions for Industry 4.0 [15, pp. 75-85]. Whereas the ap- technologies adoption to analyse their impact on the plication of machine learning, as one of the AI-based tech- maintenance of an organization. niques, in the manufacturing industry is deeply analysed in the article of Cioffi R. and others [16]. RESULTS OF RESEARCH The application of IoT improves manufacturing processes, Cognitive manufacturing extracts applicable information as machinery and equipment become more intelligent together automatically and employs analytics to get an and dynamic thanks to automation and self-optimization understanding of the manufacturing process. It robotizes [17]. The more factories and appliances are applied with reactions towards its findings and offers practical infor- the IoT, the more quantity of data is. To cope with the mation being able to steadily deliver updated knowledge large inflow of data and the complicatedness of analytics to decision-makers. computing become cognitive [18, p. 4]. Therefore, to re- Cognitive manufacturing covers four powerful applica- spond to the requirements of Industry 4.0 manufacturing tions, which are reliability and performance management transforms into cognitive manufacturing. Cognitive man- or asset performance management (APM), process and ufacturing improves businesses by using the IoT as the quality improvement, optimization of resources, and sup- base and involving advanced analytics merging with cog- ply chain optimization [10, 21, 22]. These applications are nitive technologies, as a result, it provides better indica- presented in Table 1 encompassing the information about tors for quality, efficiency, and reliability characteristics. the categories of cognitive technologies and gains of man- Techniques that fulfil and augment tasks, assist in decision ufacturers embedding technological solutions. It is found making, and help to meet targets that demand human in- that applications of cognitive technologies are classified telligence are specified as cognitive technologies. Specifi- into three main categories: product, process, and insight cally, cognitive language technologies include a group of [23] or cognitive engagement, cognitive automation, and digit technologies able to analyse, understand and output cognitive insights [24]. Though the titles differ, their human languages while interacting with machines; cogni- senses coincide. Therefore, product applications are tive machine learning provides automated analysis by al- adopted to a product or service to provide end-customer gorithms that avail of received data not necessarily envis- advantages while cognitive engagement involves cogni- aging specific programming; cognitive computer vision tive technologies application to augment the end-user ex- can extract and measure information from images even perience by proposing mass consumer personalisation at better than human vision. These technologies indicate scale. Process applications are implemented into an or- that cognitive computing can assist manufacturers and ganization’s workflow to automate or improve operations provide information to help them decide on a system of while cognitive automation is mainly used to reproduce activities. simplified mental processes. And insight applications use Manufacturing companies can benefit from smart factory cognitive technologies to reveal insights due to opera- implementation, the adoption of the production facility tional and strategic decisions while cognitive insights can that encompasses AI, robotics, analytics, big data and the develop new patterns on the base of analysed large scale M. SIRA – Efficient Practices of Cognitive Technology… 189 data sources in real-time and create additional-value in- have allowed the rail network company to achieve a cost sights. In terms of CM, we can conclude that such type of reduction of 25 per cent. Leading global manufacturer of cognitive technologies as insight prevails as cognitive sys- industrial pumps, seals, and valves and service provider of tems are capable of building relationships on base both comprehensive flow control systems has achieved a de- structured and unstructured information, conducting anal- crease in the maintenance costs and improvement of asset ysis, and generating perceptions as regards decision-mak- availability for users by up to 10 per cent; detected events ing to improve manufacturing maintenance. Process as a or predicted failures become possible in 5-6 days in ad- category cannot be minimized as manufacturing practically vance – instead of hours with an advisory application for consists of systematically connected activities, automation predictive maintenance, supported by machine learning of which can improve performance of an organization. and augmented by natural language processing [14]. APM (Table 1) improves the accuracy and maintenance of Process and quality improvement (Table 1) optimize the equipment and assets. Intelligent assets and equipment yield and productivity of manufacturing operations. Cogni- employ interconnected sensors, analytics, and cognitive ca- tive processes and operations involve the assessment of a pabilities to sense, communicate and self-diagnose any huge variety of information from workflows, context, pro- type of failures that can arise [25, p. 57]. cess and environment to quality controls, improving oper- Cognitive manufacturing applies advanced analytics and ations and decision-making [25, p.57]. Cognitive manufac- data mining techniques to decrease unnecessary down- turing tools help manufacturers to observe and understand time, and maintenance costs and enhance productivity. operational quality features more precisely and figure out Cognitive APM can not only foresee a potential issue, but it even slight problems. Manufacturers can gain an edge from can also gain information about similar cases, and relevant adopting cognitive applications that will result in increased data on how such issues have been fixed before and based revenues, reduced quality control labour costs and savings on such information it gives recommendations on how to from lower repair and warranty costs. There are two cases solve the impending failure. In the report, it is stated that listed in the report that confirm the impressive benefits of cognitive APM application has allowed an automotive man- technology adoption: a European automobile manufac- ufacturer to reduce equipment downtime by 34 per cent turer reached a 25% improvement in productivity by using and decrease equipment maintenance costs by 10 per cent predictive models while an electronics manufacturer ex- [10]. Other examples are the successful cases of the AI pre- pects a decrease in quality control labour costs by 5 to 20% dictive maintenance platform adopted by Nissan and smart [26]. Automated optical inspection systems are used by sensor and advanced machine learning analytics applied by Bosch to capture images of parts to be checked and a task- Deutsche Bahn [26, 27]. The company claims an unplanned specific software for automatically detecting defects on downtime reduction of 50% and a payback period of fewer those products [26]. Implementation of such an optical in- than 3 months. Deutsche Bahn uses smart sensors and ad- spector has made it possible to reduce total test time re- vanced machine learning analytics to reduce maintenance duction up to 45%. costs and avoid infrastructure failure. These technologies Table 1 Cognitive Manufacturing Applications Specification Prevailing application Application for CM Brief description Potentially achieved benefits type of CT Improved reliability, Cognitive APM uses the available scope performance of equipment APM of information to offer possible solutions Insight and assets, reduced costs, to alleviate upcoming issues reduced downtime Process and quality improvement involves CM process and quality instruments to observe operational Increased yield; Process and quality properties to ascertain quality regulations, provide Insight improved uptime, productivity, improvement the ability to determine the possible quality concerns revenues; reduced costs earlier and more certainly than other techniques CT enable manufactures to use resources Resource optimization Process, insight Enhanced safety, reduced costs in more efficient way CT embody data from the sensors that workers Improved worker safety Worker safety are equipped with to determine situations when em- Process and operations ployees` health threat appears in real time Energy resource Applications of cognitive and other techniques Reduced energy consumption Insight optimization allow identifying energy inefficiencies and costs Optimizing factory floor CT are used to adapt operations according Optimized planning Process, insight planning and scheduling to the carried out what-if analyses and scheduling Advantages can be achieved while adopting cognitive Fewer out-of-stock events, im- Supply chain technologies techniques gathering information from Process, insight proved on-time delivery, re- optimization structured and unstructured data roots duced inventory costs 190 Management Systems in Production Engineering 2022, Volume 30, Issue 2 Smarter resources optimization (Table 1) includes combin- with a set of options even specifying per cent of the success ing a great variety of data from different individuals, loca- of each option. As the result, the technique improves the tions, usage and expertise with cognitive insight to optimize time rate needed to accomplish the task, which influences and improve the use of resources such as labour, work- productivity and cost level. Cognitive technologies acutely force, and energy [25, p. 59]. Reduced worker downtime consider manufacturing processes and business environ- and optimised work environment, improved worker health ments to extract information important for a manufac- and productivity are the consequences achieved by a tech- turer. The procedure implies new data sources and unstruc- nology application. Cognitive technologies (CT) that enable tured data and employs advanced analytics to find out im- worker fatigue alerts and proximity monitoring used by one portant relations. Cognitive visual inspection systems can industrial product manufacturer improved the company’s carry out product checks based on images of manufactured associated safety compliance metrics by 10% [10]. Cogni- products to single out defects. Such technologies allow tive manufacturing applications can help companies to manufacturers to enhance fabric processes, and mainte- monitor and predict their energy usage and energy con- nance and diminish costs. The quality of the product is one sumption behaviour. Reduction in energy consumption due of the crucial characteristics manufacturers express con- to optimization of company’s processes can reach the rate cerns about relying on a minimum rate of defects and a high of 10 per cent [10]. rate of production accuracy as essential performance Supply chain optimization approaches (Table 1) enhance measurement. Cognitive manufacturing assists organiza- clarity and insights of gained information from structured tions in obtaining data on product quality from design and unstructured data sources to investigate the predictive through manufacturing. What is more, technologies also capability of a global production chain to minimize supply enable companies to receive information about product chain costs, disruptions and risks. quality through warranty and support programs after dis- For example, one company managed to reduce its global tribution. Such procedures boost output, reduce warranty supply chain costs by about 10% by cognitive technologies costs and support to ascertain that a customer experience embedded in an automotive Original Equipment Manufac- is positive. In addition, cognitive manufacturing prevails in turer (OEM) to resolve the suitable flow of a product using data obtained from different sources. After the infor- through the supply chain, transportation, production, and mation is analysed, there is a basis for knowledgeable sys- storage [10]. Cognitive supply chains can ensure a decrease tem creation with the possibility of constant updating and in inventories. There are two other examples when an au- learning. The main advantage of such an approach is that tomotive OEM and an industrial products manufacturer the system can make recommendations due to the under- each reduced their service parts and spare parts invento- standing of the manufacturing process and overall manu- ries, respectively, by more than 20% [10]. To give another facturing conditions. example of supply chain optimization, Continental has cre- REFERENCES ated software to predict the optimal points for tire changes [1] B.E.L.R. Flaih, D. Yuvaraj, S.K.A. Jayanthiladevi and T.S. Ku- on its fleet [26]. These predictive systems allow Continental mar, "Use Case of Artificial Intelligence in Machine Learn- to reduce its stock of tires as well as improve safety on the ing Manufacturing 4.0," International Conference on Com- road. putational Intelligence and Knowledge Economy (ICCIKE), 2019, pp. 656-659, doi: 10.1109/IC- CONCLUSIONS CIKE47802.2019.9004327. The article reveals applications for cognitive manufactur- [2] S. Wang, J. Wan, D. Li, and C. Zhang, “Implementing Smart ing, how technologies influence manufacturing processes Factory of Industrie 4.0: An Outlook,” International Journal and what their implementations result in. The theoretical of Distributed Sensor Networks, vol. 12, no. 1, p. 3159805, foundations have been developed for this purpose [28, 29]. Jan. 2016, doi: 10.1155/2016/3159805. [3] Deloitte, “2022 manufacturing industry outlook,” 2021. Today it is possible to cope with a great variety and amount Accessed: Feb. 18, 2022. [Online]. Available: of data because the level of technological development https://www2.deloitte.com/con- that assists IoT and advanced analytics is high with a possi- tent/dam/Deloitte/us/Documents/energy-resources/us- bility to scale. The technologies make it achievable for com- 2022-manufacturing-industry-outlook.pdf. panies to profit from the rapidly generated data with the [4] McKinsey & Company, “Global survey: The state of AI in help of IoT applications covering the data collected by sen- 2021|McKinsey,” www.mckinsey.com, Dec. 08, 2021. sors or unstructured data contained in other sources. Cog- https://www.mckinsey.com/business-functions/mckin- nitive technologies similar to the human brain can get a sey-analytics/our-insights/global-survey-the-state-of-ai- sense of data. The modern manufacturing era intensifies in-2021 (accessed Feb. 18, 2022). [5] A. Kwilinski and A. Kuzior. “Cognitive technologies in the targets of competitiveness, cost efficiency, flexibility, inno- management and formation of directions of the priority vation, and responsiveness, therefore it raises the signifi- development of industrial enterprises”, Management Sys- cance of the level technologies can understand data. tems in Production Engineering, 2020, 28(2), pp. 133-138. Manufacturers use cognitive technologies setting different [6] P. Fobel and A. Kuzior, “The future (Industry 4.0) is closer goals. Thus, the application of technology solutions makes than we think. Will it also be ethical?,” Proceedings of the it possible not only to solve the business problem they chal- International Conference of Computational Methods in Sci- lenge but also to enhance overall maintenance dramatically ences and Engineering 2019 (ICCMSE-2019), 2019, doi: and gain benefits. Technologies can based on the descrip- 10.1063/1.5137987. tion of the issue analyse it and provide the manufacturer M. SIRA – Efficient Practices of Cognitive Technology… 191 [7] S. Phuyal, D. Bista, and R. Bista, “Challenges, Opportunities [19] D.R. Sjödin, V. Parida, M. Leksell, and A. Petrovic, “Smart and Future Directions of Smart Manufacturing: A State of Factory Implementation and Process Innovation,” Re- Art Review,” Sustainable Futures, vol. 2, p. 100023, 2020, search-Technology Management, vol. 61, no. 5, pp. 22-31, doi: 10.1016/j.sftr.2020.100023. Sep. 2018, doi: 10.1080/08956308.2018.1471277. [8] P. Grefen, I. Vanderfeesten, K. Traganos, Z. Domagala- [20] M.F. Zaeh et al., “The Cognitive Factory,” Springer Series in Schmidt, and J. van der Vleuten, “Advancing Smart Manu- Advanced Manufacturing, pp. 355-371, doi: 10.1007/978- facturing in Europe: Experiences from Two Decades of Re- 1-84882-067-8_20. search and Innovation Projects,” Machines, vol. 10, no. 1, [21] M.S. Sheikh, “Cognitive Manufacturing in Perspective of p. 45, Jan. 2022, doi: 10.3390/machines10010045. Future Manufacturing Industries,” International Journal [9] K.-D. Thoben, S. Wiesner, and T. Wuest, “‘Industrie 4.0’ for Research in Applied Science and Engineering Technol- and Smart Manufacturing – A Review of Research Issues ogy, vol. 8, no. 10, pp. 862-866, Oct. 2020, doi: and Application Examples,” International Journal of Auto- 10.22214/ijraset.2020.32041. mation Technology, vol. 11, no. 1, pp. 4-16, Jan. 2017, doi: [22] A.V. Carvalho, A. Chouchene, T.M. Lima, and F. Charrua- 10.20965/ijat.2017.p0004. Santos, “Cognitive Manufacturing in Industry 4.0 toward [10] “Cognitive Manufacturing: An Overview and Four Applica- Cognitive Load Reduction: A Conceptual Framework,” Ap- tions that are Transforming Manufacturing Today,” 2018. plied System Innovation, vol. 3, no. 4, p. 55, Dec. 2020, doi: [Online]. Available: https://www.ibm.com/downlo- 10.3390/asi3040055. ads/cas/VDNKMWM6. [23] D. Schatsky, C. Muraskin, and M. Wagner, “Complimentary [11] I. B. M. Contributor, “IBM BrandVoice: How Cognitive article reprint Cognitive technologies The real opportuni- Computing And The IoT Can Transform Manufacturing To ties for business,” 2015. Accessed: Apr. 11, 2022. [Online]. Please Customers,” Forbes, Aug. 11, 2016. Available: https://www2.deloitte.com/tr/en/pages/tech- https://www.forbes.com/sites/ibm/2016/08/11/how- nology-media-and-telecommunications/articles/cogni- cognitive-computing-and-the-iot-can-transform-manufac- tive-technologies.html turing-to-please-customers/?sh=2dc93097272f (accessed [24] M. Lester and A. Htet, “What are cognitive technologies Feb. 19, 2022). and how are they classified? – Intelligent Automation [12] P. Moens et al., “Scalable Fleet Monitoring and Visualiza- Blog|Deloitte Australia,” Deloitte, 2019. tion for Smart Machine Maintenance and Industrial IoT Ap- https://www2.deloitte.com/au/en/blog/intelligent-auto- plications,” Sensors, vol. 20, no. 15, p. 4308, Aug. 2020, mation-blog/2019/what-are-cognitive-technologies-how- doi: 10.3390/s20154308. they-classified.html (accessed Apr. 11, 2022). [13] S. Iarovyi, J.L.M. Lastra, R. Haber, and R. del Toro, “From [25] J. Soldatos, O. Lazaro, and F. Cavadini, “The Digital artificial cognitive systems and open architectures to cog- Shopfloor: Industrial Automation in the Industry 4.0 Era,” th nitive manufacturing systems,” 2015 IEEE 13 Interna- The Digital Shopfloor: Industrial Automation in the Indus- tional Conference on Industrial Informatics (INDIN), vol. try 4.0 Era, vol. pp. 3-46, pp. 1-496, Apr. 2019, doi: pp. 1225-1232, Jul. 2015, doi: 10.13052/rp-9788770220408. 10.1109/indin.2015.7281910. [26] P. Baruchelli, F. Botto, and A. Cimatti (FBK), “End-to-end [14] Deloitte, “Exponential technologies in manufacturing,” digitized production test beds Overview on maturity of AI 2018. Accessed: Feb. 19, 2022. [Online]. Available: innovations in manufacturing,” Dec. 2020. [Online]. Avail- https://www.compete.org/storage/reports/exponen- able: https://eit.europa.eu/sites/default/files/over- tial_technologies_2018_study.pdf. view_on_maturity_of_ai_innovations_in_manufactur- [15] Kiritsis, D., Hodkiewicz, M., Lazaro, O., Lee, J., Ni, J., eds. ing_20529-d11.pdf. “Data-Driven Cognitive Manufacturing – Applications in [27] A. Kuzior and M. Staszek. Energy management in the rail- Predictive Maintenance and Zero Defect Manufacturing”. way industry: a case study of rail freight carrier in Poland. Lausanne: Frontiers Media SA. 2021, doi: 10.3389/978-2- Energies, 2021, 14 iss. 21 pp. 1-21 (art. no. 6875) 88966-583-9 https://doi.org/10.3390/en14216875 [16] R. Cioffi, M. Travaglioni, G. Piscitelli, A. Petrillo, and F. De [28] A. Kuzior, A. Kwilinski and V. Tkachenko. Sustainable de- Felice, “Artificial Intelligence and Machine Learning Appli- velopment of organizations based on the combinatorial cations in Smart Production: Progress, Trends, and Direc- model of artificial intelligence. Entrepreneurship and Su- tions,” Sustainability, vol. 12, no. 2, p. 492, Jan. 2020, doi: stainability, 2019, 7(2), 1353-1376. 10.3390/su12020492. http://doi.org/10.9770/jesi.2019.7.2(39) [17] T. Kalsoom et al., “Impact of IoT on Manufacturing Indus- [29] V. Tkachenko, A. Kuzior and A. Kwilinski. Introduction of try 4.0: A New Triangular Systematic Review,” Sustainabil- artificial intelligence tools into the training methods of en- ity, vol. 13, no. 22, p. 12506, Nov. 2021, doi: trepreneurship activities. Journal of Entrepreneurship Edu- 10.3390/su132212506. cation, 2019, 22(6), pp. 1-10. Retrieved from [18] S. Bonnaud, C. Didier, and A. Kohler, “Industry 4.0 and Cog- https://www.abacademies.org/articles/Introduction-of- nitive Manufacturing Architecture Patterns, Use Cases and artificialintelligence-tools-1528-2651-22-6-477.pdf IBM Solutions,” Sep. 2019. [Online]. Available: https://www.ibm.com/downloads/cas/M8J5BA6R. Mariya Sira ORCID ID: 0000-0002-9970-1549 Silesian University of Technology e-mail: Mariya.Sira@polsl.pl http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Management Systems in Production Engineering de Gruyter

Efficient Practices of Cognitive Technology Application for Smart Manufacturing

Management Systems in Production Engineering , Volume 30 (2): 5 – Jun 1, 2022

Loading next page...
 
/lp/de-gruyter/efficient-practices-of-cognitive-technology-application-for-smart-lK130EueBW
Publisher
de Gruyter
Copyright
© 2022 Mariya Sira, published by Sciendo
eISSN
2450-5781
DOI
10.2478/mspe-2022-0023
Publisher site
See Article on Publisher Site

Abstract

Cognitive manufacturing (CM) provides for the merging of sensor-based information, advanced analytics, and cognitive technologies, mainly machine learning in the context of Industry 4.0. Manufacturers apply cognitive technologies to review current business metrics, solve essential business problems, generate new value in their manufacturing data and improve quality. The article investigates four powerful applications for cognitive manu- facturing and their influence on a company`s maintenance. The study aims to observe kinds of cognitive technol- ogy applications for smart manufacturing, distinguish their prospective gains for manufacturers and provide suc- cessful examples of their adoption. The analysis is based on the literature and report review. Assessment of the cases of technology adoption proves that cognitive manufacturing provides both enhanced knowledge manage- ment and helps organizations improve fundamental business measurements, such as productivity, product relia- bility, quality, safety, and yield while reducing downtime and lowering costs. Key words: cognitive manufacturing, cognitive technology, Industry 4.0 INTRODUCTION adopted AI use cases within manufacturing [4, 5]. Across Innovative scientific tendencies, the emergence of manu- manufacturing respondents, they report higher levels of facturing 4.0 and cognitive technologies impact manufac- cost decreases from AI adoption in more than one-and- turing procedures and effectiveness. The fundamental one-half times in the fiscal year 2020 compared to the fis- idea of Industry 4.0 is to adopt the Internet of Things (IoT), cal year 2019 with a dramatic increase of percentage rep- artificial intelligence (AI), and cyber-physical systems thus resenting a decrease of more than 20 per cent while rev- manufacturing process and business process are ex- enue increase remains almost at the same level [4]. tremely integrated to provide flexible and efficient pro- Considering the cognitive technology to be a part of the duction [1, 2]. Manufacturing continues to change as or- larger area of AI and a dramatic increase in technologies ganizations equip their factory with technologies and ap- adoption, cognitive technology application for smart man- ply complex data analytics to improve performance, prod- ufacturing is worth examining. Hence the article aims to uct quality, and the way perception of manufacturing in- investigate four applications for cognitive manufacturing, formation is found and responded to. what areas of manufacturing they can refer to, what are According to the 2022 manufacturing industry outlook by advantages for manufacturers for embedding CT into pro- Deloitte, an increase in investment in artificial intelligence duction and give examples confirming the success of cog- technologies is expected at a compound annual growth nitive technology application. rate above 20% through 2025 [3]. A global transformation LITERATURE REVIEW is in progress to equip manufacturing with AI. Discrete The use of technologies in different fields under Industry manufacturing ranks among the three industries with the 4.0, such as robotics and artificial intelligence, poses a highest levels of contribution in AI including quality man- new requirement [6]. The approach of cognitive manufac- agement and automated preventive maintenance use turing is consonant with such concepts as smart manufac- cases [3]. turing [7], and Industry 4.0 [3, 8, 9]. The authors [8, 9] fig- In the McKinsey Global Survey "The State of AI in 2021" ure out Industry 4.0 as the most essential part of the predictive maintenance and yield, energy, and/or Fourth Industrial Revolution with smart manufacturing throughput optimization are defined to be the most © 2022 Author(s). This is an open access article licensed under the Creative Commons BY 4.0 (https://creativecommons.org/licenses/by/4.0/ 188 Management Systems in Production Engineering 2022, Volume 30, Issue 2 being part of it, overlapping concepts, namely the indus- IoT, with an increase in process efficiency, lower opera- trial internet, the industrial internet of things and intelli- tional costs, increased safety and sustainability, and in- gent manufacturing. creased product quality [19]. Manufacturers adopt cogni- Two major technologies that are pushing Industry 4.0 are tive technologies to detail manufacturing processes and IoT and analytics [10, 11]. IoT allows factories to improve business environments aiming at obtaining information their instrumented and interconnected status with the suited for further manufacturing processes digitizing and ability to provide gained relevant data together from the optimization. Targets for cost-effectiveness, quality, prod- manufacturing environment in real-time. Moens and oth- uct reliability, cost, and time efficiency force organizations ers [12] provide an Industrial Internet of Things (IIoT) sys- to search for ways how to improve their manufacturing tem that includes smart maintenance solutions. They re- processes. veal the robustness and scaling of solutions as well as the The concept of cognitive manufacturing is relevant to the availability of well-trained machine learning models in ability to deal with complex modern production systems case of defect recognition. Industrial IoT establishes new and have a quick reaction to unforeseeable issues in pro- features in industrial automation systems which are de- duction, planning and control systems. Thus, the Cogni- termined by improving system robustness and flexibility tive Factory is assumed to be flexible, adaptable, reliable, [13]. and efficient in various momentary situations [20]. Analytics stands for determining templates in the data, a METHODOLOGY pattern of equipment approach and a forecast of possible The methodology of this article was based on a three-step breakdowns. It implies a variety of statistical techniques approach. The first step included preliminary research on including pattern recognition, text analytics, cluster anal- the concept of cognitive manufacturing and obtaining a ysis, factor analysis, multivariate modelling, multiple re- deeper knowledge of the research area. The second stage gression, forecasting, machine learning, simulation, and consisted of the examination of academic literature and neural networks [14]. The research of Rousopoulou V. and report review on the applications for cognitive manufac- others introduce cognitive analytics, self-and autono- turing leading to a restraint of the scope of a theoretical mous-learned system bearing predictive maintenance so- foundation. The last one covered case studies of cognitive lutions for Industry 4.0 [15, pp. 75-85]. Whereas the ap- technologies adoption to analyse their impact on the plication of machine learning, as one of the AI-based tech- maintenance of an organization. niques, in the manufacturing industry is deeply analysed in the article of Cioffi R. and others [16]. RESULTS OF RESEARCH The application of IoT improves manufacturing processes, Cognitive manufacturing extracts applicable information as machinery and equipment become more intelligent together automatically and employs analytics to get an and dynamic thanks to automation and self-optimization understanding of the manufacturing process. It robotizes [17]. The more factories and appliances are applied with reactions towards its findings and offers practical infor- the IoT, the more quantity of data is. To cope with the mation being able to steadily deliver updated knowledge large inflow of data and the complicatedness of analytics to decision-makers. computing become cognitive [18, p. 4]. Therefore, to re- Cognitive manufacturing covers four powerful applica- spond to the requirements of Industry 4.0 manufacturing tions, which are reliability and performance management transforms into cognitive manufacturing. Cognitive man- or asset performance management (APM), process and ufacturing improves businesses by using the IoT as the quality improvement, optimization of resources, and sup- base and involving advanced analytics merging with cog- ply chain optimization [10, 21, 22]. These applications are nitive technologies, as a result, it provides better indica- presented in Table 1 encompassing the information about tors for quality, efficiency, and reliability characteristics. the categories of cognitive technologies and gains of man- Techniques that fulfil and augment tasks, assist in decision ufacturers embedding technological solutions. It is found making, and help to meet targets that demand human in- that applications of cognitive technologies are classified telligence are specified as cognitive technologies. Specifi- into three main categories: product, process, and insight cally, cognitive language technologies include a group of [23] or cognitive engagement, cognitive automation, and digit technologies able to analyse, understand and output cognitive insights [24]. Though the titles differ, their human languages while interacting with machines; cogni- senses coincide. Therefore, product applications are tive machine learning provides automated analysis by al- adopted to a product or service to provide end-customer gorithms that avail of received data not necessarily envis- advantages while cognitive engagement involves cogni- aging specific programming; cognitive computer vision tive technologies application to augment the end-user ex- can extract and measure information from images even perience by proposing mass consumer personalisation at better than human vision. These technologies indicate scale. Process applications are implemented into an or- that cognitive computing can assist manufacturers and ganization’s workflow to automate or improve operations provide information to help them decide on a system of while cognitive automation is mainly used to reproduce activities. simplified mental processes. And insight applications use Manufacturing companies can benefit from smart factory cognitive technologies to reveal insights due to opera- implementation, the adoption of the production facility tional and strategic decisions while cognitive insights can that encompasses AI, robotics, analytics, big data and the develop new patterns on the base of analysed large scale M. SIRA – Efficient Practices of Cognitive Technology… 189 data sources in real-time and create additional-value in- have allowed the rail network company to achieve a cost sights. In terms of CM, we can conclude that such type of reduction of 25 per cent. Leading global manufacturer of cognitive technologies as insight prevails as cognitive sys- industrial pumps, seals, and valves and service provider of tems are capable of building relationships on base both comprehensive flow control systems has achieved a de- structured and unstructured information, conducting anal- crease in the maintenance costs and improvement of asset ysis, and generating perceptions as regards decision-mak- availability for users by up to 10 per cent; detected events ing to improve manufacturing maintenance. Process as a or predicted failures become possible in 5-6 days in ad- category cannot be minimized as manufacturing practically vance – instead of hours with an advisory application for consists of systematically connected activities, automation predictive maintenance, supported by machine learning of which can improve performance of an organization. and augmented by natural language processing [14]. APM (Table 1) improves the accuracy and maintenance of Process and quality improvement (Table 1) optimize the equipment and assets. Intelligent assets and equipment yield and productivity of manufacturing operations. Cogni- employ interconnected sensors, analytics, and cognitive ca- tive processes and operations involve the assessment of a pabilities to sense, communicate and self-diagnose any huge variety of information from workflows, context, pro- type of failures that can arise [25, p. 57]. cess and environment to quality controls, improving oper- Cognitive manufacturing applies advanced analytics and ations and decision-making [25, p.57]. Cognitive manufac- data mining techniques to decrease unnecessary down- turing tools help manufacturers to observe and understand time, and maintenance costs and enhance productivity. operational quality features more precisely and figure out Cognitive APM can not only foresee a potential issue, but it even slight problems. Manufacturers can gain an edge from can also gain information about similar cases, and relevant adopting cognitive applications that will result in increased data on how such issues have been fixed before and based revenues, reduced quality control labour costs and savings on such information it gives recommendations on how to from lower repair and warranty costs. There are two cases solve the impending failure. In the report, it is stated that listed in the report that confirm the impressive benefits of cognitive APM application has allowed an automotive man- technology adoption: a European automobile manufac- ufacturer to reduce equipment downtime by 34 per cent turer reached a 25% improvement in productivity by using and decrease equipment maintenance costs by 10 per cent predictive models while an electronics manufacturer ex- [10]. Other examples are the successful cases of the AI pre- pects a decrease in quality control labour costs by 5 to 20% dictive maintenance platform adopted by Nissan and smart [26]. Automated optical inspection systems are used by sensor and advanced machine learning analytics applied by Bosch to capture images of parts to be checked and a task- Deutsche Bahn [26, 27]. The company claims an unplanned specific software for automatically detecting defects on downtime reduction of 50% and a payback period of fewer those products [26]. Implementation of such an optical in- than 3 months. Deutsche Bahn uses smart sensors and ad- spector has made it possible to reduce total test time re- vanced machine learning analytics to reduce maintenance duction up to 45%. costs and avoid infrastructure failure. These technologies Table 1 Cognitive Manufacturing Applications Specification Prevailing application Application for CM Brief description Potentially achieved benefits type of CT Improved reliability, Cognitive APM uses the available scope performance of equipment APM of information to offer possible solutions Insight and assets, reduced costs, to alleviate upcoming issues reduced downtime Process and quality improvement involves CM process and quality instruments to observe operational Increased yield; Process and quality properties to ascertain quality regulations, provide Insight improved uptime, productivity, improvement the ability to determine the possible quality concerns revenues; reduced costs earlier and more certainly than other techniques CT enable manufactures to use resources Resource optimization Process, insight Enhanced safety, reduced costs in more efficient way CT embody data from the sensors that workers Improved worker safety Worker safety are equipped with to determine situations when em- Process and operations ployees` health threat appears in real time Energy resource Applications of cognitive and other techniques Reduced energy consumption Insight optimization allow identifying energy inefficiencies and costs Optimizing factory floor CT are used to adapt operations according Optimized planning Process, insight planning and scheduling to the carried out what-if analyses and scheduling Advantages can be achieved while adopting cognitive Fewer out-of-stock events, im- Supply chain technologies techniques gathering information from Process, insight proved on-time delivery, re- optimization structured and unstructured data roots duced inventory costs 190 Management Systems in Production Engineering 2022, Volume 30, Issue 2 Smarter resources optimization (Table 1) includes combin- with a set of options even specifying per cent of the success ing a great variety of data from different individuals, loca- of each option. As the result, the technique improves the tions, usage and expertise with cognitive insight to optimize time rate needed to accomplish the task, which influences and improve the use of resources such as labour, work- productivity and cost level. Cognitive technologies acutely force, and energy [25, p. 59]. Reduced worker downtime consider manufacturing processes and business environ- and optimised work environment, improved worker health ments to extract information important for a manufac- and productivity are the consequences achieved by a tech- turer. The procedure implies new data sources and unstruc- nology application. Cognitive technologies (CT) that enable tured data and employs advanced analytics to find out im- worker fatigue alerts and proximity monitoring used by one portant relations. Cognitive visual inspection systems can industrial product manufacturer improved the company’s carry out product checks based on images of manufactured associated safety compliance metrics by 10% [10]. Cogni- products to single out defects. Such technologies allow tive manufacturing applications can help companies to manufacturers to enhance fabric processes, and mainte- monitor and predict their energy usage and energy con- nance and diminish costs. The quality of the product is one sumption behaviour. Reduction in energy consumption due of the crucial characteristics manufacturers express con- to optimization of company’s processes can reach the rate cerns about relying on a minimum rate of defects and a high of 10 per cent [10]. rate of production accuracy as essential performance Supply chain optimization approaches (Table 1) enhance measurement. Cognitive manufacturing assists organiza- clarity and insights of gained information from structured tions in obtaining data on product quality from design and unstructured data sources to investigate the predictive through manufacturing. What is more, technologies also capability of a global production chain to minimize supply enable companies to receive information about product chain costs, disruptions and risks. quality through warranty and support programs after dis- For example, one company managed to reduce its global tribution. Such procedures boost output, reduce warranty supply chain costs by about 10% by cognitive technologies costs and support to ascertain that a customer experience embedded in an automotive Original Equipment Manufac- is positive. In addition, cognitive manufacturing prevails in turer (OEM) to resolve the suitable flow of a product using data obtained from different sources. After the infor- through the supply chain, transportation, production, and mation is analysed, there is a basis for knowledgeable sys- storage [10]. Cognitive supply chains can ensure a decrease tem creation with the possibility of constant updating and in inventories. There are two other examples when an au- learning. The main advantage of such an approach is that tomotive OEM and an industrial products manufacturer the system can make recommendations due to the under- each reduced their service parts and spare parts invento- standing of the manufacturing process and overall manu- ries, respectively, by more than 20% [10]. To give another facturing conditions. example of supply chain optimization, Continental has cre- REFERENCES ated software to predict the optimal points for tire changes [1] B.E.L.R. Flaih, D. Yuvaraj, S.K.A. Jayanthiladevi and T.S. Ku- on its fleet [26]. These predictive systems allow Continental mar, "Use Case of Artificial Intelligence in Machine Learn- to reduce its stock of tires as well as improve safety on the ing Manufacturing 4.0," International Conference on Com- road. putational Intelligence and Knowledge Economy (ICCIKE), 2019, pp. 656-659, doi: 10.1109/IC- CONCLUSIONS CIKE47802.2019.9004327. The article reveals applications for cognitive manufactur- [2] S. Wang, J. Wan, D. Li, and C. Zhang, “Implementing Smart ing, how technologies influence manufacturing processes Factory of Industrie 4.0: An Outlook,” International Journal and what their implementations result in. The theoretical of Distributed Sensor Networks, vol. 12, no. 1, p. 3159805, foundations have been developed for this purpose [28, 29]. Jan. 2016, doi: 10.1155/2016/3159805. [3] Deloitte, “2022 manufacturing industry outlook,” 2021. Today it is possible to cope with a great variety and amount Accessed: Feb. 18, 2022. [Online]. Available: of data because the level of technological development https://www2.deloitte.com/con- that assists IoT and advanced analytics is high with a possi- tent/dam/Deloitte/us/Documents/energy-resources/us- bility to scale. The technologies make it achievable for com- 2022-manufacturing-industry-outlook.pdf. panies to profit from the rapidly generated data with the [4] McKinsey & Company, “Global survey: The state of AI in help of IoT applications covering the data collected by sen- 2021|McKinsey,” www.mckinsey.com, Dec. 08, 2021. sors or unstructured data contained in other sources. Cog- https://www.mckinsey.com/business-functions/mckin- nitive technologies similar to the human brain can get a sey-analytics/our-insights/global-survey-the-state-of-ai- sense of data. The modern manufacturing era intensifies in-2021 (accessed Feb. 18, 2022). [5] A. Kwilinski and A. Kuzior. “Cognitive technologies in the targets of competitiveness, cost efficiency, flexibility, inno- management and formation of directions of the priority vation, and responsiveness, therefore it raises the signifi- development of industrial enterprises”, Management Sys- cance of the level technologies can understand data. tems in Production Engineering, 2020, 28(2), pp. 133-138. Manufacturers use cognitive technologies setting different [6] P. Fobel and A. Kuzior, “The future (Industry 4.0) is closer goals. Thus, the application of technology solutions makes than we think. Will it also be ethical?,” Proceedings of the it possible not only to solve the business problem they chal- International Conference of Computational Methods in Sci- lenge but also to enhance overall maintenance dramatically ences and Engineering 2019 (ICCMSE-2019), 2019, doi: and gain benefits. Technologies can based on the descrip- 10.1063/1.5137987. tion of the issue analyse it and provide the manufacturer M. SIRA – Efficient Practices of Cognitive Technology… 191 [7] S. Phuyal, D. Bista, and R. Bista, “Challenges, Opportunities [19] D.R. Sjödin, V. Parida, M. Leksell, and A. Petrovic, “Smart and Future Directions of Smart Manufacturing: A State of Factory Implementation and Process Innovation,” Re- Art Review,” Sustainable Futures, vol. 2, p. 100023, 2020, search-Technology Management, vol. 61, no. 5, pp. 22-31, doi: 10.1016/j.sftr.2020.100023. Sep. 2018, doi: 10.1080/08956308.2018.1471277. [8] P. Grefen, I. Vanderfeesten, K. Traganos, Z. Domagala- [20] M.F. Zaeh et al., “The Cognitive Factory,” Springer Series in Schmidt, and J. van der Vleuten, “Advancing Smart Manu- Advanced Manufacturing, pp. 355-371, doi: 10.1007/978- facturing in Europe: Experiences from Two Decades of Re- 1-84882-067-8_20. search and Innovation Projects,” Machines, vol. 10, no. 1, [21] M.S. Sheikh, “Cognitive Manufacturing in Perspective of p. 45, Jan. 2022, doi: 10.3390/machines10010045. Future Manufacturing Industries,” International Journal [9] K.-D. Thoben, S. Wiesner, and T. Wuest, “‘Industrie 4.0’ for Research in Applied Science and Engineering Technol- and Smart Manufacturing – A Review of Research Issues ogy, vol. 8, no. 10, pp. 862-866, Oct. 2020, doi: and Application Examples,” International Journal of Auto- 10.22214/ijraset.2020.32041. mation Technology, vol. 11, no. 1, pp. 4-16, Jan. 2017, doi: [22] A.V. Carvalho, A. Chouchene, T.M. Lima, and F. Charrua- 10.20965/ijat.2017.p0004. Santos, “Cognitive Manufacturing in Industry 4.0 toward [10] “Cognitive Manufacturing: An Overview and Four Applica- Cognitive Load Reduction: A Conceptual Framework,” Ap- tions that are Transforming Manufacturing Today,” 2018. plied System Innovation, vol. 3, no. 4, p. 55, Dec. 2020, doi: [Online]. Available: https://www.ibm.com/downlo- 10.3390/asi3040055. ads/cas/VDNKMWM6. [23] D. Schatsky, C. Muraskin, and M. Wagner, “Complimentary [11] I. B. M. Contributor, “IBM BrandVoice: How Cognitive article reprint Cognitive technologies The real opportuni- Computing And The IoT Can Transform Manufacturing To ties for business,” 2015. Accessed: Apr. 11, 2022. [Online]. Please Customers,” Forbes, Aug. 11, 2016. Available: https://www2.deloitte.com/tr/en/pages/tech- https://www.forbes.com/sites/ibm/2016/08/11/how- nology-media-and-telecommunications/articles/cogni- cognitive-computing-and-the-iot-can-transform-manufac- tive-technologies.html turing-to-please-customers/?sh=2dc93097272f (accessed [24] M. Lester and A. Htet, “What are cognitive technologies Feb. 19, 2022). and how are they classified? – Intelligent Automation [12] P. Moens et al., “Scalable Fleet Monitoring and Visualiza- Blog|Deloitte Australia,” Deloitte, 2019. tion for Smart Machine Maintenance and Industrial IoT Ap- https://www2.deloitte.com/au/en/blog/intelligent-auto- plications,” Sensors, vol. 20, no. 15, p. 4308, Aug. 2020, mation-blog/2019/what-are-cognitive-technologies-how- doi: 10.3390/s20154308. they-classified.html (accessed Apr. 11, 2022). [13] S. Iarovyi, J.L.M. Lastra, R. Haber, and R. del Toro, “From [25] J. Soldatos, O. Lazaro, and F. Cavadini, “The Digital artificial cognitive systems and open architectures to cog- Shopfloor: Industrial Automation in the Industry 4.0 Era,” th nitive manufacturing systems,” 2015 IEEE 13 Interna- The Digital Shopfloor: Industrial Automation in the Indus- tional Conference on Industrial Informatics (INDIN), vol. try 4.0 Era, vol. pp. 3-46, pp. 1-496, Apr. 2019, doi: pp. 1225-1232, Jul. 2015, doi: 10.13052/rp-9788770220408. 10.1109/indin.2015.7281910. [26] P. Baruchelli, F. Botto, and A. Cimatti (FBK), “End-to-end [14] Deloitte, “Exponential technologies in manufacturing,” digitized production test beds Overview on maturity of AI 2018. Accessed: Feb. 19, 2022. [Online]. Available: innovations in manufacturing,” Dec. 2020. [Online]. Avail- https://www.compete.org/storage/reports/exponen- able: https://eit.europa.eu/sites/default/files/over- tial_technologies_2018_study.pdf. view_on_maturity_of_ai_innovations_in_manufactur- [15] Kiritsis, D., Hodkiewicz, M., Lazaro, O., Lee, J., Ni, J., eds. ing_20529-d11.pdf. “Data-Driven Cognitive Manufacturing – Applications in [27] A. Kuzior and M. Staszek. Energy management in the rail- Predictive Maintenance and Zero Defect Manufacturing”. way industry: a case study of rail freight carrier in Poland. Lausanne: Frontiers Media SA. 2021, doi: 10.3389/978-2- Energies, 2021, 14 iss. 21 pp. 1-21 (art. no. 6875) 88966-583-9 https://doi.org/10.3390/en14216875 [16] R. Cioffi, M. Travaglioni, G. Piscitelli, A. Petrillo, and F. De [28] A. Kuzior, A. Kwilinski and V. Tkachenko. Sustainable de- Felice, “Artificial Intelligence and Machine Learning Appli- velopment of organizations based on the combinatorial cations in Smart Production: Progress, Trends, and Direc- model of artificial intelligence. Entrepreneurship and Su- tions,” Sustainability, vol. 12, no. 2, p. 492, Jan. 2020, doi: stainability, 2019, 7(2), 1353-1376. 10.3390/su12020492. http://doi.org/10.9770/jesi.2019.7.2(39) [17] T. Kalsoom et al., “Impact of IoT on Manufacturing Indus- [29] V. Tkachenko, A. Kuzior and A. Kwilinski. Introduction of try 4.0: A New Triangular Systematic Review,” Sustainabil- artificial intelligence tools into the training methods of en- ity, vol. 13, no. 22, p. 12506, Nov. 2021, doi: trepreneurship activities. Journal of Entrepreneurship Edu- 10.3390/su132212506. cation, 2019, 22(6), pp. 1-10. Retrieved from [18] S. Bonnaud, C. Didier, and A. Kohler, “Industry 4.0 and Cog- https://www.abacademies.org/articles/Introduction-of- nitive Manufacturing Architecture Patterns, Use Cases and artificialintelligence-tools-1528-2651-22-6-477.pdf IBM Solutions,” Sep. 2019. [Online]. Available: https://www.ibm.com/downloads/cas/M8J5BA6R. Mariya Sira ORCID ID: 0000-0002-9970-1549 Silesian University of Technology e-mail: Mariya.Sira@polsl.pl

Journal

Management Systems in Production Engineeringde Gruyter

Published: Jun 1, 2022

Keywords: cognitive manufacturing; cognitive technology; Industry 4.0

There are no references for this article.