Abstract
The use of artificial intelligence (AI) is today’s dominating technological trend across all industries. With the maturing of deep learning and other data-driven techniques, AI has over the last decade become an essential component for an increasing number of products and services. In parallel to this development, technological advances have been accelerating the production of novel AI models from large-scale datasets. This global phenomenon has been driving the need for an efficient industrialized approach to develop, manage and maintain AI models at large scale. Such an approach is provided by the state-of-the-art operational concept termed AI Factory, which refers to an infrastructure for AI models and implements the idea of AI as a Service (AIaaS). Moreover, it ensures performance, transparency and reproducibility of AI models at any point in the continuous AI development process. This concept, however, does not only require new technologies and architectures, but also new job roles. Here, we discuss current trends, outline requirements and identify success factors for AI Factories. We conclude with recommendations for their successful use in practice as well as perspectives on future developments.
About the authors
Dr. Wolfgang Hildesheim is a high-energy physicist by training and spent several years conducting research at the CERN and DESY research centers. In 2007, he took over the management of the Automotive, Aerospace and High Tech unit at IBM. From 2009, he led IBM’s Big Data Industry Solution Business in Europe and supported companies in the development of data-driven business models using advanced analytics. Since 2012, he has been co-responsible for the foundation as well as the development of IBM’s business unit Watson, Data Science & Artificial Intelligence in Europe in various technical and sales roles often with a focus on Germany, Austria and Switzerland. Wolfgang Hildesheim is a member of the board of BitKom’s Artificial Intelligence Working Group. Further, he is actively involved in the standardization of artificial intelligence, for example, as a member of the steering group of the German Standardization Roadmap Artificial Intelligence and as a delegate to the European committee CEN-CENELEC JTC 21 Artificial Intelligence for the establishment of a classification scheme for the homogeneous description of artificial intelligence within all member countries of the European Union.
Taras Holoyad is an electrical engineer and works in the field of artificial intelligence standardization at the Federal Network Agency in Mainz. After graduating from the Technical University of Braunschweig, he calculated electrical machines for road vehicles in the automotive industry before joining the most important German regulatory authority. Taras Holoyad has been contributing to the first and second editions of the German Institute for Standardization’s (DIN) Artificial Intelligence Standardization Roadmap, where he was involved in developing concepts relevant to the global regulation of artificial intelligence. He has also been researching quality characteristics of RGB and IR cameras on traffic signals for AI-based analysis of road traffic in collaboration with the city of Wiesbaden. In cooperation with Nordhausen University of Applied Sciences, he is also working on the identification of semantic similarities between documents with a focus on patent and legal texts.
Dr. Thomas Schmid is a W1 Professor for Digital Research Methods in Medicine at Martin Luther University Halle-Wittenberg and has been developing data-driven artificial intelligence techniques and systems for more than a decade. After studying bioinformatics in Tübingen, Gaborone (Botswana) and Berlin, he first has been research on the use of neural networks for biomedical applications in Leipzig and the USA in cooperation with Charité Berlin. After receiving his PhD in computer science at Leipzig University in 2018, Thomas Schmid established the Machine Learning group there, before co-founding the computer science department of Lancaster University in Leipzig. Since 2021, he has been chairing and developing the annual Leipzig Symposium on Intelligent Systems (LEISYS). His current research interests include biomedical applications of machine learning and the development of hybrid artificial intelligence approaches. Together with Wolfgang Hildesheim and Taras Holoyad, he authored the book Künstliche Intelligenz managen und verstehen (Beuth Publishers, 2023).
References
[1] J. P. Simon, “Artificial intelligence: scope, players, markets and geography,” Digit. Policy, Regul. Gov., vol. 21, no. 3, pp. 208–237, 2019. https://doi.org/10.1108/dprg-08-2018-0039.Search in Google Scholar
[2] A. Kordon, “Applied artificial intelligence-based systems as competitive advantage,” in 2020 IEEE 10th International Conference on Intelligent Systems (IS), Varna, Bulgaria, 2020, pp. 6–18.10.1109/IS48319.2020.9200097Search in Google Scholar
[3] de Oca, I. Moreno-Montes, H. A. Reijers, and A. Rodríguez-Morffi, “A systematic literature review of studies on business process modeling quality,” Inf. Softw. Technol., vol. 58, pp. 187–205, 2015. https://doi.org/10.1016/j.infsof.2014.07.011.Search in Google Scholar
[4] P. Bocciarelli, A. D’Ambrogio, A. Mastromattei, et al.., “Business process modeling and simulation: state of the art and MSaaS opportunities,” in Proceedings of the Summer Simulation Multi-Conference, x, x, 2017.Search in Google Scholar
[5] V. R. S. Kumar, A. Khamis, S. Fiorini, et al.., “Ontologies for industry 4.0,” Knowl. Eng. Rev., vol. 34, p. e17, 2019. https://doi.org/10.1017/s0269888919000109.Search in Google Scholar
[6] H. Panetto and L. Whitman, “Knowledge engineering for enterprise integration, interoperability and networking: theory and applications,” Data Knowl. Eng., vol. 105, pp. 1–4, 2016. https://doi.org/10.1016/j.datak.2016.05.001.Search in Google Scholar
[7] B. Marr, Artificial Intelligence in Practice: How 50 Successful Companies Used AI and Machine Learning to Solve Problems, John Wiley & Sons, 2019.Search in Google Scholar
[8] J. Aastha, D. Shah, and P. Churi, “A review on business intelligence systems using artificial intelligence,” in Computational Vision and Bio-Inspired Computing. ICCVBIC 2019. Advances in Intelligent Systems and Computing 1108, S. Smys, J. Tavares, V. Balas, and A. Iliyasu, Eds., 2019.Search in Google Scholar
[9] I. H. Sarker, “AI-based modeling: techniques, applications and research issues towards automation, intelligent and smart systems,” SN Comput. Sci., vol. 3, p. 158, 2022. https://doi.org/10.1007/s42979-022-01043-x.Search in Google Scholar PubMed PubMed Central
[10] T. Schmid, “Modelle und Modellierung,” in Künstliche Intelligenz managen und verstehen - Der Praxis-Wegweiser für Entscheidungsträger, Entwickler und Regulierer, T. Schmid, W. Hildesheim, and T. Holoyad, Eds., Berlin, Beuth Publishers, 2023.Search in Google Scholar
[11] E. Hechler, M. Oberhofer, and T. Schaeck, Deploying AI in the Enterprise, Berkeley, CA, apress, 2020.10.1007/978-1-4842-6206-1Search in Google Scholar
[12] E. Hechler, M. Weihrauch, and Y. Wu, “Evolution of data architecture,” in Data Fabric and Data Mesh Approaches with AI, Berkeley, CA, Apress, 2023.Search in Google Scholar
[13] M. Johnson, R. Jain, P. Brennan-Tonetta, et al.., “Impact of big data and artificial intelligence on industry: developing a workforce roadmap for a data driven economy,” Glob. J. Flex. Syst. Manag., vol. 22, no. 3, pp. 197–217, 2021. https://doi.org/10.1007/s40171-021-00272-y.Search in Google Scholar
[14] A. Ashta and H. Herrmann, “Artificial intelligence and fintech: an overview of opportunities and risks for banking, investments, and microfinance,” Strateg. Change, vol. 30, pp. 211–222, 2021. https://doi.org/10.1002/jsc.2404.Search in Google Scholar
[15] R. Clarke, “Principles and business processes for responsible AI,” Comput. Law Secur. Rev., vol. 35, no. 4, pp. 410–422, 2019. https://doi.org/10.1016/j.clsr.2019.04.007.Search in Google Scholar
[16] E. Kurshan, H. Shen, and J. Chen, “Towards self-regulating AI: challenges and opportunities of AI model governance in financial services,” in Proceedings of the First ACM International Conference on AI in Finance, 2020, pp. 1–8.10.1145/3383455.3422564Search in Google Scholar
[17] EU Commission, White Paper on Artificial Intelligence – A European Approach to Excellence and Trust, Brussels, EU Commission, 2020.Search in Google Scholar
[18] European Commission, Proposal for a Regulation of the European Parliament and of the Council – Laying Down Harmonised Rules on Artificial Intelligence, Brussels, EU Commission, 2021.Search in Google Scholar
[19] M. Cherradi and A. EL Haddadi, “Data lakes: a survey paper,” in Innovations in Smart Cities Applications Volume 5. SCA 2021. Lecture Notes in Networks and Systems, vol. 393, M. Ben Ahmed, A. A. Boudhir, I. R. Karas, V. Jain, and S. Mellouli, Eds., Cham, Springer, 2022.10.1007/978-3-030-94191-8_66Search in Google Scholar
[20] E. Hechler, M. Weihrauch, and Y. Wu, “Terminology: data fabric and data mesh,” in Data Fabric and Data Mesh Approaches with AI, Berkeley, CA, Apress, 2023.10.1007/978-1-4842-9253-2Search in Google Scholar
[21] D. E. O’Leary, “Artificial intelligence and big data,” IEEE Intell. Syst., vol. 28, no. 2, pp. 96–99, 2013. https://doi.org/10.1109/mis.2013.39.Search in Google Scholar
[22] D. Sculley, G. Holt, D. Golovin, et al.., “Hidden technical debt in machine learning systems,” Adv. Neural Inf. Process., vol. 28, pp. 2503–2511, 2015.Search in Google Scholar
[23] A. Paleyes, R. G. Urma, and N. D. Lawrence, “Challenges in deploying machine learning: a survey of case studies,” ACM Comput. Surv., vol. 55, no. 6, pp. 1–29, 2022. https://doi.org/10.1145/3533378.Search in Google Scholar
[24] M. Iansiti and K. R. Lakhani, “The AI factory,” in Competing in the Age of AI: Strategy and Leadership when Algorithms and Networks Run the World, M. Iansiti and K. R. Lakhani, Eds., Harvard Business Press, 2020.Search in Google Scholar
[25] N. Yuhanna, A. Katz, A. Lozada, and K. Pierpont, The 15 Providers that Matter Most and How They Stack up. The Forrester WaveTM: Enterprise Data Fabric, Q2 2022, June 23, 2022.Search in Google Scholar
[26] F. Wenzel and A. Haag, “Die AI Factory der ERGO,” in Künstliche Intelligenz und Data Science in Theorie und Praxis, A. Gillhuber, G. Kauermann, and W. Hauner, Eds., Berlin, Heidelberg, Springer Spektrum, 2023.10.1007/978-3-662-66278-6_15Search in Google Scholar
[27] J. Siderska, “Robotic Process Automation — a driver of digital transformation?” Eng. Manag. Prod. Serv., vol. 12, no. 2, pp. 21–31, 2020. https://doi.org/10.2478/emj-2020-0009.Search in Google Scholar
[28] P. Buxmann and H. Schmidt, Künstliche Intelligenz, Berlin, Heidelberg, Springer Gabler, 2021.10.1007/978-3-662-61794-6Search in Google Scholar
[29] S. Bouyon, Recent Trends in EU Home Ownership. ECRI Commentary No. 10691, Brussels, Belgium, Centre for European Policy Studies, 2015.Search in Google Scholar
[30] D. W. Schrotenboer, The Impact of Artificial Intelligence along the Customer Journey: A Systematic Literature Review, Bachelor thesis, University of Twente, 2019.Search in Google Scholar
[31] S. Anagnoste, “Robotic Automation Process–The operating system for the digital enterprise,” Proc. Int. Conf. Bus. Excell., vol. 12, no. 1, pp. 54–69, 2018. https://doi.org/10.2478/picbe-2018-0007.Search in Google Scholar
[32] P. Hofmann, C. Samp, and N. Urbach, “Robotic process automation,” Electron. Mark., vol. 30, no. 1, pp. 99–106, 2020. https://doi.org/10.1007/s12525-019-00365-8.Search in Google Scholar
[33] M. Meesters, P. Heck, and A. Serebrenik, “What is an AI engineer? An empirical analysis of job ads in The Netherlands,” in Proceedings of the 1st International Conference on AI Engineering: Software Engineering for AI, 2022, pp. 136–144.10.1145/3522664.3528594Search in Google Scholar
[34] D. Kreuzberger, N. Kühl, and S. Hirschl, “Machine learning operations (MLOps): overview, definition, and architecture,” IEEE Access, vol. 11, pp. 31866–31879, 2023. https://doi.org/10.1109/ACCESS.2023.3262138.Search in Google Scholar
[35] D. De Silva and D. Alahakoon, “An artificial intelligence life cycle: from conception to production,” Patterns, vol. 3, p. 100489, 2022. https://doi.org/10.1016/j.patter.2022.100489.Search in Google Scholar PubMed PubMed Central
[36] D. Brosseau, S. Ebrahim, C. Handscomb, and S. Thaker, The Journey to an Agile Organization, Report, McKinsey & Company, 2019.Search in Google Scholar
[37] J. S. Saltz and N. W. Grady, “The ambiguity of data science team roles and the need for a data science workforce framework,” in 2017 IEEE International Conference on Big Data (Big Data), Boston, MA, USA, 2017, pp. 2355–2361.10.1109/BigData.2017.8258190Search in Google Scholar
[38] J. K. Hentzen, A. Hoffmann, R. Dolan, and E. Pala, “Artificial intelligence in customer-facing financial services: a systematic literature review and agenda for future research,” Int. J. Bank Mark., vol. 40, pp. 1299–1336, 2021. https://doi.org/10.1108/ijbm-09-2021-0417.Search in Google Scholar
[39] O. H. Fares, I. Butt, and S. H. M. Lee, “Utilization of artificial intelligence in the banking sector: a systematic literature review,” J. Financ. Serv. Mark., pp. 1–18, 2022, https://doi.org/10.1057/s41264-022-00176-7.Search in Google Scholar
[40] T. Granlund, A. Kopponen, V. Stirbu, L. Myllyaho, and T. Mikkonen, “MLOps challenges in multi-organization setup: experiences from two real-world cases,” in 2021 IEEE/ACM 1st Workshop on AI Engineering – Software Engineering for AI (WAIN), Madrid, Spain, 2021, pp. 82–88.10.1109/WAIN52551.2021.00019Search in Google Scholar
[41] M. Mahadevan, “Jes borland,” in Data Professionals at Work, Berkeley, CA, Apress, 2018.10.1007/978-1-4842-3967-4Search in Google Scholar
[42] T. Schmid, W. Hildesheim, T. Holoyad, and K. Schumacher, “The AI methods, capabilities and criticality grid – a three-dimensional classification scheme for artificial intelligence applications,” Kunstl. Intell., vol. 35, pp. 425–440, 2021. https://doi.org/10.1007/s13218-021-00736-4.Search in Google Scholar
[43] S. Lins, K. D. Pandl, H. Teigeler, S. Thiebes, C. Bayer, and A. Sunyaev, “Artificial intelligence as a service,” Bus. Inf. Syst. Eng., vol. 63, pp. 441–456, 2021. https://doi.org/10.1007/s12599-021-00708-w.Search in Google Scholar
[44] A. Bogdanov, A. Degtyarev, N. Shchegoleva, V. Korkhov, and V. Khvatov, “Big data virtualization: why and how,” CEUR Workshop Proc., vol. 2679, pp. 11–21, 2020.Search in Google Scholar
[45] A. Krogh, “What are artificial neural networks?” Nat. Biotechnol., vol. 26, no. 2, pp. 195–197, 2008. https://doi.org/10.1038/nbt1386.Search in Google Scholar PubMed
[46] D. Pessach and E. Shmueli, “A review on fairness in machine learning,” ACM Comput. Surv, vol. 55, no. 3, pp. 1–44, 2022. https://doi.org/10.1145/3494672.Search in Google Scholar
[47] J. Lu, A. Liu, F. Dong, F. Gu, J. Gama, and G. Zhang, “Learning under concept drift: a review,” IEEE Trans. Knowl. Data Eng., vol. 31, no. 12, pp. 2346–2363, 2018.10.1109/TKDE.2018.2876857Search in Google Scholar
[48] V. Belle and I. Papantonis, “Principles and practice of explainable machine learning,” Front. Big Data, vol. 4, p. 688969, 2021. https://doi.org/10.3389/fdata.2021.688969.Search in Google Scholar PubMed PubMed Central
[49] Gesellschaft für Informatik, Exam AI: Abschlussbericht Exam AI – KI Testing und Auditing, Report, 2021. Available at: https://testing-ai.gi.de/fileadmin/PR/Testing-AI/Abschlussbericht_ExamAI_-_KI_Testing_und_Auditing.pdf [accessed: Sep. 15, 2023].Search in Google Scholar
[50] Bundesministeriums für Wirtschaft und Energie, Technologieprogramm KI-Innovationswettbewerb des Bundesministeriums für Wirtschaft und Energie: Erklärbare KI, Report, 2021. Available at: https://www.digitale-technologien.de/DT/Redaktion/DE/Downloads/Publikation/KI-Inno/2021/Studie_Erklaerbare_KI.pdf;jsessionid=1C6B11B563FE0B546B8249DD51218659?__blob=publicationFile&v=17 [accessed: Sep. 15, 2023].Search in Google Scholar
[51] B. Stanton and T. Jensen, Trust and Artificial Intelligence, NIST Interagency/Internal Report (NISTIR), Gaithersburg, MD, National Institute of Standards and Technology, 2021. Available at: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=931087 [accessed: Sep. 15, 2023].Search in Google Scholar
[52] J. Mökander, M. Axente, F. Casolari, and L. Floridi, “Conformity assessments and post-market monitoring: a guide to the role of auditing in the proposed European AI regulation,” Minds Mach., vol. 32, pp. 241–268, 2022. https://doi.org/10.1007/s11023-021-09577-4.Search in Google Scholar PubMed PubMed Central
[53] Datenethikkommission der Bundesregierung, Datenethikkommission der Bundesregierung: Gutachten, Berlin, 2019.Search in Google Scholar
[54] Deutsches Institut für Normung.Normungsroadmap Künstliche Intelligenz, Deutsches Institut für Normung, 2020.Search in Google Scholar
© 2023 Walter de Gruyter GmbH, Berlin/Boston
