How to create an copilot for the shop floor
Building a copilot may seem as simple as connecting a chatbot to a database but, in an industrial setting, reaching the level of knowledge and input required makes it a far more complex task. Ensuring both human users and the copilot itself have everything they need to succeed is a challenge in its own right while the requirements of a digital twin to provide a true brain to the copilot is both a key necessity and an additional challenge to overcome.
In a recent podcast, host Conor Peick was joined by Theo Papadopoulos, senior consultant and head of the Metaverse Lab at Siemens to explore what it takes to allow copilots and humans to work hand in hand, both from the perspective of the user and the AI assistant.
Check out the full episode here or keep reading for a summary of the highlights of that conversation.
Understanding the needs of copilots and users
To make copilots useful, the key is, as always, user experience. Not only must the system provide useful information, such as ‘what is this error?’ or ‘what is the history of this part?’ it must do so in a convent way that makes sense to the user as well. Theo describes it as a problem of making sure the user is excited to use the tool because if not, it’ll be obsolete in a matter of ‘hours or days’.
Bringing copilots to the shop floor through the lens of VR is both a practical and perilous approach to tackling this problem. On the one hand, VR glasses offer an excellent method of presenting information to both users and AI yet, in the past, VR technology has struggled to reach widespread adoption. In order to assure success with users, not only must the copilot itself be a useful tool but the method of accessing it must also be just as seamless to use.
On the other hand, copilots themselves must also be properly equipped to address user requests. An important baseline for this, especially on the shop floor, is to know the state of any given machine. Without this minimal level of information, it’s impossible to even begin to answer the question of how do I fix this problem? Beyond that, copilots must know not only the current state of the machine but its history as well, in order to help identify systemic issues and past fixes. Finally, understanding intent is equally important; the ability to interpret from a users request what they actually want to achieve and provide appropriate guidance in that direction. Combining these factors together, alongside sufficiently robust information, are what makes copilots on the shop floor truly valuable.
The importance of the Digital Twin
If smart glasses are the eyes, then the Digital Twin is the brain behind an Industrial Copilot as Theo explains. The Digital Twin not only provides a powerful repository of knowledge, but also important links and context between different pieces of information as well. These semantic connections are important to understanding how to correctly approach problems, the interactions of different machines and systems, as well as being a repository of both up to date and historical data.
The Digital Twin is also an important element in allowing AI to answer ‘what if?’ questions. In many scenarios the user may ask a copilot how a proposed change will effect the system as a whole. In order to answer this question, the AI requires a place to safely simulate those changes, checking how they directly effect the target machine as well as the greater impact of that change on production systems as a whole. By going to the Digital Twin with this question, the copilot can safely test changes and report the results entirely in the digital world, without the need for physical experimentation or the risk of inaccurate results.
Building an Industrial Copilot for industry is a daunting task, yet, with the right merger of technologies and careful attention to detail, one with a great deal to offer on completion. Merging cutting edge AI and wearable technology with the trusted power of the Digital Twin can bring a lot to workers on the shop floor, bringing all the perks of digitalization right to their fingertips.
Siemens Digital Industries Software helps organizations of all sizes digitally transform using software, hardware and services from the Siemens Xcelerator business platform. Siemens’ software and the comprehensive digital twin enable companies to optimize their design, engineering and manufacturing processes to turn today’s ideas into the sustainable products of the future. From chips to entire systems, from product to process, across all industries. Siemens Digital Industries Software – Accelerating transformation.
