While artificial intelligence (AI) has played a pivotal role in driving the shift to digitalization
in manufacturing, and companies have adopted AI to automate and collect data for many industrial devices, the overall adoption of AI is still slow in this sector.
The true competitive advantage will only be realized by companies that embrace a combination of advanced technologies and AI, particularly a form of AI that more closely mimics human behavior.
Consider how AI previously required supervised learning with well-structured and perfectly labelled data sets. The next generation of AI will require less pre-labelling because the networks will essentially teach themselves by discovering patterns in the data independently, or semi-independently. These ‘unsupervised-learning’ approaches that are currently being used for image recognition are now being applied in other fields, such as quality control in manufacturing, medical imaging, and autonomous driving.
By minimizing the need for feeding perfect datasets to the AI system, industrial organizations will now have a faster, more efficient way to make use of AI. At the same time, many of the biggest advances expected in industrial AI are related to changes in the way data is managed, generated, represented and shared. For example:
• Contextual data and simulations: We will see AI applied to data sets created and organized in new ways to enhance insights and understanding. Examples include knowledge graphs, which capture the meaning of (and relationships between) items in diverse data sets, and digital twins, which provide detailed digital representations and simulations of real systems, assets and processes.
• Embedded AI and big-picture insights: Internet of Things (IoT) and edge technologies are giving rise to diverse machine-generated data sets that can support new levels of situational awareness and real-time insights in the cloud or directly in the field.
• Data from beyond the walls: Improved protocols and technologies for sharing data (e.g. standardized semantic description of data sets and points) could lead to greater exchange of data between organizations, and this could help to develop AI models that simultaneously draw from the data of suppliers, partners, regulators, customers and perhaps even competitors.
While organizations are historically very protective of their data and processes and have strict rules on sharing information externally, the next generation AI breakthroughs will depend on smart data exchanges between organizations. New innovations such as blockchain-exchange systems can automate the sharing of data and ensure only specific, identified organizations receive the approved sections of data. As benefits become more apparent from smart-data exchanges, organizations will likely become more open to the automation of data exchange between various types of organizations.
As digitalization continues to grow in the industry, manufacturers are using AI to make sense of vast IoT datasets, all while edge computing is helping to embed automated intelligence into the world. These twin trends and new capabilities in the next generation of industrial AI will drive greater predictive power, more autonomous machines and richer simulations for more educated decisions, safer facilities, and more productivity.
Alessandra Da Silva is head of artificial intelligence and edge computing deployment with Siemens Industry