Machines have helped make us exceedingly efficient.
When Sir Richard Arkwright threw his beliefs behind the very first cotton spinning machine, he did so due to his faith in a crossed-mix of two disciplines: watchmaking and reed-making. The result was a mechanical spinning machine that put the then rather manual textiles industry into a spin.
However, later iterations of Arkwright’s machines grew in complexity and scale, and needed to be run by something more powerful. They became entire factories – some of the first factories in the world.
Arkwright also continued to combine these factories with new technologies of the time. At first, it was with river-driven water wheels, but later he brought in steam power once it came to fruition. Soon, his factories, powered by steam and water, were the only realistic way to mass produce textiles.
Many dispute whether Arkwright, now widely considered the ‘father of the modern industrial factory system’, truly invented the machines and factories that completely changed the cottage industry. Nevertheless, it is clear that without the combination of technologies of the time, the first industrial revolution would not have taken off as it did.
Likewise, today’s factories face a similar tipping point. Technologies such as Internet connectivity, artificial intelligence (AI), cloud computing and of course an explosion of data are propelling us into another era of industrial productivity.
In today’s factories, it is common to see in efficient automated assembly lines, computerized or digitalized processes, even robotic arms that perform with speed and precision that humans will never be able to replicate. However, we suspect we will see iterative improvements in these in the near future.
Future flexible factories
Assembly lines need not be fixed. Factories used to be laid out in such a way that they can only produce or assemble a single type of product. Even with product variants, such as different types of bread in a bread factory, it often requires lengthy and costly preparation and re-configuration time.
Today, designers and engineers are rethinking factory layouts and bringing in more flexibility onto factory floors. Increasingly, factory floors are being laid out in a modular fashion instead of the previously popular long assembly lines, allowing for workgroups to be changed out and replaced quickly during re-configuration. This flexibility also allows factories to modify production arrangements in the event of unexpected events, such as machine failure.
Additionally, mobile robots are being introduced to factory floors to move material and products, and even whole machines, to where they are needed. With sensors and built-in AI, some of these robots are capable of learning the layout of buildings, and automatically map out the most efficient and safest routes to take when transporting goods and material. They can also be programmed to collaborate with each other and work together as a mobile fleet for maximum efficiency.
Dynamic data-driven factories
Today’s information age and the abundance of data that comes with you also means that smart factories are able to operate more intelligently. Data brings about a more dynamic production capability as data-driven factories can be designed to produce at the optimum level automatically. Hence, new machines should be designed with “Intelligent Automation” from the onset while for legacy machines, they should be retrofitted to be able to capture critical intelligent information.
For instance, future factories can potentially prioritise production processes based on information like fluctuating raw material cost, demand and orders received, utility costs and so on. Smart factories can align output according to these factors. They may also be able to find a balance on the production of several different products, and dynamically switch among them according to cost efficiency and direct demand.
Apart from consuming this data, smart factories may also be generating their own data -- data that can be used to further increase productivity. Manufacturers can switch to predictive rather than reactive maintenance by collecting data from the factory floors using “smart” sensors in IIoT enabled machines. They can optimise maintenance schedules, while collecting insights into strain, performance and potential bottlenecks of the overall production line.
Constructive collaborative factories
The human role will always be needed even in the smartest of factories. Critical thinking and complex assembly processes in some factories means that the best solution is for humans and robots to work together.
Smart factories are equipped with sensors and automated systems that enable safe direct contact between humans and robots which needs “Interactive Automation” technology to realise this safe interaction. This is in contrast with many of today’s factories where many robotic arms are isolated in cages and programmed to halt activities immediate as a safety precaution whenever humans enter a ‘production zone’.
The smart collaborative environment will allow robots that co-work with humans to aid the human workers on specific tasks, such as heavy load bearing or tasks that require contact with hazardous material. Another example are robots armed with special sensors that continuously monitor the state of health of their human co-workers, and alerting or stopping them from overexerting themselves or overstepping the bounds of what is considered safe.
The possibilities for smart factories are pretty much limitless as technology advances. Designers and engineers will continue to seek resource efficiency, drive productivity and look into ways where factories can function at their optimum level taking into consideration factors like operational costs, partner requirements and market demand. As Arkwright combined his machines with cutting edge technologies of his time to spark the first industrial revolution, so too can we use the technologies of today to build smart factories of tomorrow.
This article was first published on Questex, on 15 October 2018. Information is correct at the time of publication.