Data Is the Key for Systems-Level Robotics Transformation
The robotics revolution is creating big possibilities in terms of productivity. And as AI continues developing at a rapid pace, the unlock to seeing expectations translated into reality is a systems-level data model.
Robots have captured our imagination since Josef Čapek coined the term in 1920. The idea was always about machines toiling in place of humans. Today, with AI exploding, investors and boards expect a productivity revolution in manufacturing and logistics with robots leading the way.
Data is the key to meeting those expectations.
Task-Level Data Is Necessary but Not Sufficient
Industrial robot installations are rising fast, but traditional automation is still purpose-built for narrow applications. Coding and setup are a barrier, with engineers balancing upfront cost and time against longer-term payback based primarily on labor savings.
Using robots in place of humans in traditional automation requires data that describes the operation of machinery and details of the workpiece. This is normally high precision, high volume, low variance, structured data that is isolated from higher-level product data (ie, bill of materials) and supply chain data (ie, orders, work instructions, identity, etc). De-bottlenecking with task-level automation will find incremental savings indefinitely but with declining returns over time.
System-Level Data Unlocks the End-to-End Power of AI in Manufacturing and Logistics
Pulling up from a task to a system level still requires machine control data but also a closed-loop view of the business purpose driving large-scale automation plans. This may include supply and material options, people and regulatory considerations, and product lifecycle accountability. The center of this systems-level data model is either “product” for manufacturing-oriented robotics systems or “customer promise” for logistics robotics systems.
Toyota, for example, has invested billions in manufacturing robotics using a systems-thinking approach that isn’t about eliminating labor but instead improving quality, safety, and flexibility in auto assembly. This application of robotics and AI includes classical quality inspection, AGVs for moving parts around, and automated welding operations, as well as human work analytics meant to identify dangerous or stressful process steps.
The Toyota Research Institute has also developed a genAI breakthrough called Diffusion Policy based on Large Behavior Models (LBMs), which are the robot equivalent of Large Language Models (LLMs) for chatbots. The engineering advantage is much faster training and greater dexterity. At a system level, the business win is more agile production for the world’s second-largest automaker.
JD.com’s logistics arm has also utilized automation and robotics, emerging as a unicorn in the e-commerce space. With thousands of robots and capabilities that span AGVs, robotic arms, and automated storage and retrieval systems, the company has bolstered its pioneering beyond just China.
Since building its own logistics in 2017, the parent company has grown 127% in annual revenues while JD Logistics has seen an astronomical 360% increase in annual revenues. JD Logistics CEO Yu Rui describes the business as the “AWS of China’s supply chain,” referring to the systems-level thinking Amazon embodies in the form of working backward to build the capabilities needed to reach a specific goal.
It is not necessary to populate every element of the conceptual data models we share above to make progress toward operational performance improvement. It is important, however, to have a data strategy that includes at least some of them since evolving AI technologies demand more comprehensive data about business problems and the intended outcomes of system-level robotic transformations.
System-Level ROI Beats Task-Level ROI
Future-proofing robotics investments means thinking through data sources and uses holistically across the business. For example, a digital twin that maintains software version control, maintenance history, and operating data for each unit sold can give end customers more value. Tesla, John Deere, and Rolls-Royce are examples of manufacturers applying this principle.
For others like Novartis, Beiersdorf, and Levi’s, the emphasis may be on upstream data like raw material sources and production accountability for business cases based on consumer safety or responsible sourcing. Automation in these cases requires data that enables lot-level traceability to protect brand reputation and enhance equity.
The bottom line is that robots are advancing fast, and AI is speeding things up. Without the right data strategy, breakthrough results are beyond reach.