Leading-edge technologies that can make an impact
Search for technology trends that are relevant to the manufacturing sector and you’ll find dozens of different articles on the subject.
There is no definitive list, no “must have” technology, that manufacturers need to embrace but, neither is there any shortage of technologies designed to help manufacturers deal with the unique challenges wrought from the digital world in which we live.
It only serves to underscore our longstanding view, reflected in the guidance we provide our clients, that manufacturers need to carefully consider which technology they want to purchase and deploy and which challenge or challenges they want to solve before making any decisions.
With that in mind, here are five technologies that are worth considering today and that are also pointing the way over the next several years.
The name says it all. Two identical objects but one is physically real, while the other is a digital version, a computer-generated model. This model is exact in all respects to its physical other and is based on the acquisition and analysis of data from the physical object.
Data drives manufacturing today.
Thanks to millions of cheap sensors embedded in machines and finished products and through analyzing the data gathered from those sensors, a manufacturer can understand how its product performs in the real world. Does it break? Does it show early signs of wear? Does it perform well in conditions of extreme temperature fluctuation?
Information like this can be used to create computer models which can then be “bench-tested” in a digital environment every bit as real as a physical environment.
The net result is more rapid operational and product improvement, increased reliability and fewer recalls or warranty claims.
5G and Edge Computing
Let’s begin with edge computing. Some argue that this is a throwback to the days of on-premises data storage. After all, the cloud, they argue, makes storing large amounts of data far more scalable, secure and cheaper and eliminates the need for expensive hardware like servers. Edge, they claim, simply parks data on-site.
However, manufacturers are producing products like refrigerators or industrial equipment with embedded sensors that must be linked to the IoT or Industrial IoT (IIoT). The sensors contain their own limited data gathering and analysis capabilities and share the data back to the manufacturer via the IoT.
An industrial air conditioning unit in a meat plant gathers and interprets data on its internal workings on-site and relays it to the manufacturer’s service center. The data doesn’t have to be sent to a public cloud first.
The point of edge computing is to allow for computing capability “at the scene” as much as possible which helps to reduce latency – the amount of time required to send data from one point to the next. A perfect example of computing at the scene is the smartphone in your hand. Another is the autonomous automobile.
As the number of edge endpoints (like that fridge, smartphone or self-driving car) around the world increases exponentially, thanks to cheaper and cheaper sensors and rapidly expanding data production, so, too, does the need to link them efficiently with each other and with back end services.
5G, the latest cellular technology, is the answer. It will have the bandwidth (upload capacities of 10 gigabytes per second; download capacities of 20gps), the lowest latency levels (below 5 milliseconds) and massive scalability.
AI and Machine Learning
Machine learning is really a subset of artificial intelligence (AI) and refers to computer-controlled equipment that has the built-in capacity to learn through repeated actions, thanks to algorithms in the software. This is not a new technology, per se, but its uses and usefulness keep expanding as more and more data is produced in the manufacturing sector.
AI only works when data is generated and analyzed. So, a manufacturer may receive an analytics-driven report indicating a spike in sales of its products. AI will take that further, analyzing enormous amounts of data to determine why sales spiked and offer predictions of what could happen if the price was discounted or how unit costs could be lowered if volumes were increased.
Since manufacturers are always looking for ways to balance sales forecasts with production capacity, AI offers a very real tool to help.
Robotic Process Automation
Robots in manufacturing are usually associated with the value stream – what’s actually transpiring on the factory floor; vehicles are assembled largely by robotic machines along a production line, for example, that often employs more robots than humans. This application of robotics has been a major contributor to reduced unit costs and a dramatic reduction in product deficiency which, in turn, has lowered warranty costs.
Robotic Process Automation (RPA) moves the concept of robotics from the factory floor to the front office.
RPA is actually software designed to fully automate human activities that are manual, rule-based and repeatable. Today, there is a prevalence of ‘chat boxes’, which allow a customer to interact with a software program to solve a relatively straightforward problem like resetting a password or solving other customer service issues. This is an example of the usefulness of RPA as is automating data entry or processing standard transactions.
RPA can free up humans previously charged with dealing with routine tasks for other assignments.
One key with RPA is that it does not necessitate changes to existing business systems. Some process automation systems can only function through the use of Application Programming Interfaces or APIs, which are third-party software programs.
Manufacturing As A Service (MaaS)
What cloud technology has ushered in is an efficient way to obtain latest-version software. For consumers buying a new home computer, there is no more disc installation – because there are no more discs in the box. Instead, operating systems are downloaded under licence.
The beauty of this method is that updates are downloaded just as easily and their cost is rolled into a renewable subscription agreement. Hence, the term ‘software-as-a-service’.
Manufacturing-as-a-Service takes this model further; as customer demands for low volume or custom parts continue to grow, many manufacturers find they can’t keep up.
This has given rise to MaaS, where a business can upload the specs for a custom part and receive quotes from various MaaS 3D manufacturers in minutes. The entire manufacturing process for that business speeds up without diverting financial resources to the acquisition of on-site 3D printers or the time-consuming production of custom or low volume products.
As a result, both overall labor and unit costs remain under control.
So, five examples of technology available now for manufacturers to employ. This is far from a comprehensive list which makes the process of deciding what to choose even more onerous for most manufacturing enterprises.
Gerent has been providing guidance and expertise for more than ten years to clients around the world. We are not simply a Salesforce implementation company. We partner with our clients and we provide the direction needed to help them navigate what is a seemingly endless world of technology.