Internet of Things & Predictive Analytics
What is it?
The Internet of Things (IoT) has been called the cornerstone of 4IR, and can be described as a network of sensors in products, that produces data on its use. More specific to manufacturing is the Industrial Internet of Things (IIoT) in which sensors are embedded in machine parts, robots/cobots, end products, and other manufacturing equipment at different stages of the manufacturing process, which enables them to send and receive data. It allows for a closer integration of machines, robots, factory equipment, and a company’s IT systems, as well as closer supply chain and customer integration.
Predictive analytics is closely linked to IoT and can be described as the use of data gathered by sensors based on statistical algorithms, modelling and machine learning techniques to make predictions about the likelihood of future outcomes.
What are the possible benefits?
IoT sensors can gather operational machine data as well as strategic data. This data can be used throughout the supply chain, and during the lifetime of the product. Analysis of this data can give companies specific and accurate information in real time about their operations and therefore lead to process as well as product optimisation and predictive maintenance when products or machinery are faulty and require repair. It can also improve yield and speed up production.
The data gathered can then be applied to many other aspects of 4IR technology such as blockchain, AR and cobots. For example, the data gathered from IoT sensors will often contain important company data that may be best stored in a blockchain, enabling only those with permission to access the information.
Predictive analytics has great potential to manufacturers in that it can assist in anticipating needs throughout a factory by automating much of the analysis processes. Predictive analysis can assist in quality control, demand forecast, machine utilisation, preventative maintenance and detecting fraud. The predictions are used to determine issues such as how long machines can be used on average before needing maintenance and what products and how much of the products should be produced at different times of the year, amongst other uses.
Internet of Things and Predictive Analytics in Practice
One UK casting manufacturer uses sensors to enable machine monitoring of when the machine is running or idle and reasons why the machine may be stopped. this in turn helps the manufacturer to improve processes, machine maintenance and have real time information from several sites.
A print manufacturer knows the usage times and consumable levels of its machines and sends replacement for consumables without the need for ordering. As part of this they are also able to manage upgrades on their own schedule and based on the data from all machines predict when maintenance will be required and schedule this. This allows them to sell the ‘service’ rather than the product to the end customer.
We want to hear your stories of 4IR investment. Share your experiences more widely and help manufacturers like you by emailing [email protected] to tell us your story.