From the Industrial Revolution through the Gulf War, the state of robots has progressed rapidly. This rapid growth has been accompanied by companies developing sentient machines which are more creative and compassionate than humans.

While scientists are harnessing the power of robots to improve their own ability to work and conduct humanitarian projects, data from a study published in Nature suggests that production robots might be a much more resource-intensive form of automation than traditional robot manufacturing.

The study suggests that automated production jobs could pose a significant threat to human jobs as robots grow. The study, conducted by researchers at the University of Guelph, concluded that automated production jobs provide better job security than traditional robot manufacturing jobs.

With the rise of robot manufacturing, many manufacturers are facing the prospect of layoffs. Promoting technological advances in robotics and computerization has improved the productivity of labour intensive industrial processes, which is present in many large businesses, such as financial institutions. Given the relative efficiency of robots, increasing automation has encouraged innovation in engineering and manufacturing, reducing redundancy.

As the production economy improves, workers are brought into the loop to regulate production processes. This has made large companies able to plan their production lines in advance, avoid major disruptions to the supply chain and promote efficiencies.

However, increasing automation could also reduce jobs, so researchers, led by Guelph Professors Catharine Asma and Dr. Leo W. Bonfils, created a statistical method to identify the production robotics based on sensor inputs. The system was designed to evaluate whether manual labour is required to produce parts rather than robots.

The researchers highlighted that automation already reduces labour intensive manufacturing of non-automotive goods, such as farm labour and power tools. However, during peak production periods, robots used at sensitive parts often require manual labour.

“All you need is 10,000 sensors, and you can do things like automate these jobs,” said Mr. Bonfils. “There is a robot being produced, and so when you have a sudden technological surge, people have to stop doing the work for a while and look at each other and say, ‘How can I do this?’”

For example, he argued, someone who has a short time to produce a robot component could switch back to labour so as to keep up with requirements. “Technicians spend a lot of time there learning how to find sensors and custom tune them and reassemble,” he said. “But people who need to do repetitive work are much more inclined to take a different path to it.”

The findings suggest that the report of reducing repetitive labour for machines may prove to be short-sighted. Robots may play a role in reducing labour intensive labour, but they cannot replace it completely. While robots have been developed in a position to relieve repetitive labour, it is highly unlikely that they can replace it completely.

Still, the study suggests that AI could assist traditional manufacturing when humans were unable to oversee operation. By focusing less on manual skills, machine learning could provide an additional level of productivity to the production sector.

“AI is much easier to use in the manufacturing sector now than it was a few years ago,” said Dr. Bonfils. “When humans are working and robots are being produced, machines are literally working for the sake of work. It would be wonderful if we could change the nature of work to support productivity in machine manufacturing.”

While Mr. Bonfils estimates that 20 per cent of the human labour force can be saved simply by automating manual jobs, he warned that the reduction in labour, such as manual labour for robots, may actually be a strain on human resources.

“When it comes to the whole question of how AI affects labour, it is a lot more complicated than it appears,” he said. “If you want to engineer robots that work in isolation from the rest of the population, you’re going to get more output per person and more employees. It doesn’t mean you’re going to decrease the size of the workforce. It’s just that you’re not going to be able to recruit the same number of people to be on the job.”

However, Mr. Bonfils noted that automation could help eliminate out of sight the need for humans. “If robots were human and you were able to harness their skill sets in the lowest most specific job, automation could help reduce the need for humans.”

The report is published in the journal Nature.