Summary of the Key Points
This report brings a very sobering reality to the currently hot humanoid robotics industry: While all manufacturers are boasting about the operating hours, efficiency, and practical outcomes of their products, these figures are essentially just scattered “attendance records” that prove the robots have been turned on and used. However, none of these cases provide a comprehensive account of the actual costs—such as how much was spent on purchasing the robot, on debugging and maintaining it, how much work it has completed, how long it was out of service, and how much labor it has saved for the employer. In other words, the industry only has an “attendance sheet” and lacks a “payroll sheet” that would demonstrate whether robots are truly more cost-effective than hiring employees.
The claim by the American manufacturer Agility that they can recoup their investment in 1.1 years is purely based on idealized assumptions and is not a result derived from actual customer experiences. The industry has yet to overcome the final hurdle of transforming from demonstration toys into profitable production tools.
---
Detailed Explanation in Plain Language
1. The Promised 1.1-Year Payback Is a Masked Mathematical Trick
Many people might be tempted by the idea of earning back their investment in just over a year by spending $200,000 on a robot. However, upon closer examination of the assumptions behind Agility’s figures, it becomes clear that this is more of a sales pitch for investors. When calculating the payback, Agility excludes the one-time deployment cost of $20,000 and the annual software maintenance fee of $36,000, focusing only on the cost of the robot itself ($200,000) and comparing it to the annual labor value of $190,000. The company’s ideal scenario assumes the robot works 120 hours per week, which is three times the workload of a typical employee, running non-stop for 17 hours a day for five years without any downtime or errors, with each minute of work worth $30.5. Such a robot doesn’t exist in reality; this figure is merely a fabricated story to support Agility’s $2.5 billion valuation for its IPO.
2. Don’t Be Misled by Operating Hours
The first six months a robot is in use are often a period of trial and error, during which it’s not actually productive. Manufacturers’ claims of “X million hours of operation” exaggerate the situation significantly. For example, a robot from Xingdongjiyuan that could move 600 parcels per hour in the lab may only move 100 parcels per hour in a real logistics environment due to poor lighting and irregular parcel shapes, and it may often malfunction. Engineers have to work overtime to adjust parameters and monitor its performance, much like training a new employee. During this period, the company may even incur additional costs to support the robot.
BMW’s claimed 1,250 hours of operation represent the total running time of multiple robots in the project, not the actual effective working time of a single robot. These figures include debugging, downtime, and inefficiencies, and thus do not accurately reflect the value created by the robots.
3. The Value of Work Varies Dramatically Depending on the Context
The value of a robot’s work cannot be simply compared across different scenarios. Moving parcels in a logistics center, tightening high-precision screws in a car factory, or providing patient care in a hospital are all vastly different tasks. Just as a one-hour shift for a delivery driver or a programmer does not equate in value, the same one-hour of work for a robot cannot be directly compared. Even if a robot’s efficiency improves after six months in a logistics setting, it cannot be transferred to another context without re-calibrating its parameters, rendering previous performance meaningless.
4. The Missing “Real Payroll Sheet” Is the Trigger for Industry Growth
The fact that manufacturers are now sharing data on actual usage is a positive sign of progress, as it indicates that robots are being integrated into real production environments. The industry is moving beyond the “toy stage.” The “real payroll sheet” that the industry needs—i.e., a comprehensive analysis of costs and benefits—will be determined by the first customer to provide concrete evidence of the robot’s value. If such a case emerges, demand will surge, and manufacturers will not need to promote their products aggressively. Conversely, if no such case emerges in the next two to three years, the current high valuations and expectations will prove to be unsustainable, leading to a major reshuffle in the industry.