With the fourth industrial revolution, or better known as Industry 4.0, beginning to show signs of incorporation in today’s manufacturing industry, there have been many bombastic terms spawned: Internet of Things (IoT), Industry 4.0, Smart Manufacturing, Big Data, Digital Factory. And truth be told, not all manufacturers around the world are clear of what exactly these are yet.
Defining Industry 4.0
In brief, the Industry 4.0 that originated in Germany envisions the adoption and execution of a ‘Smart Factory’, where all manufacturing processes would ideally be automated, less human labour would be required, operation efficiencies would be improved and customisation of products to be mass produced will be made available.
In doing this, the Industry 4.0 revolves around six design principles:
- Interoperability: humans and machines will be able to communicate via the IoT—humans will be able to operate machines remotely and at a higher machine-to-human ratio.
- Virtualisation: the ability to view the entire factory and manufacturing processes virtually.
- Decentralisation: the ability of machines to ‘think’ on their own for smart decision-making.
- Real-time Capability: With the smarter functions of machines, more data would also be generated and the capability of machines to collect, analyse and provide derived insights in real-time would be an ideal asset.
- Service orientation: the ability to offer services through the Internet of Services
- Modularity: the flexibility of Smart Factories to accommodate changing requirements through replacing or expanding individual modules.
A Big Data Mess
As aforementioned, with Industry 4.0 coming into play and the more sophisticated machines available, manufacturers today are producing more data daily than they did some 10 years ago such as sensor data, camera images, digital gauge data, etc.
In theory, all this data that is churned out can be used to measure things. However, managing and making sense of this data—termed ‘big data’—in practice is a challenge, because while all these data can be generated or gathered, manufacturers are posed with the complexity of what to use this data for or how to even use it.
To solve this problem, data management companies are competing to create the leading data analytics programme that can best serve manufacturers of the industry. One area in which these programmes are rapidly developing in is the Industry Internet of Things (IIoT).
The programmes developed under this serve to capture, collate and distribute data to the company’s personnel, while customising it totally so that the relevant personnel would receive only data that they would need, and in convenient ways such as through their personal devices—smart phones, tablets, or laptops.
To store the unquantifiable amounts of information pouring out from the machines, cloud storage is currently popularly being employed as a repository for large unstructured data sets of machine process variables, tooling and workpiece information, due to the economic viability of leveraging the cloud. Being on the internet that can be accessed virtually from any location, data can be tackled and analysed from anywhere and at any point of time.
But how, one may ask, is using this data beneficial? How does it affect one’s manufacturing operations? Is the data important?
How Data Can Be Used
Without a doubt, analysing the data produced can benefit manufacturers in a myriad of ways. For one, data on processing times or energy consumption for example, let a manufacturer with quantifiable evidence of the efficiency of the process or machine, and if an alternative is required. Processes or machines in this way can be optimised once the direction for improvement has been identified through the use of data.
According to Sal Spada, research director, ARC Advisory Group, machine tool data can also predict the quality of a production workpiece. With real-time process parameters that can be extracted from operating machines, production machinery process models can be developed and continuously improved upon to guarantee the quality of production workpieces.
Also, with the increased connectivity between digitally-controlled machines, manufacturers can additionally collect a wide array of real-time process parameters. This data can then be used to develop accurate mathematical models of machine behaviour under innumerable operating conditions that are difficult to set up in a laboratory test bed, he added. This would then allow manufacturers to virtually monitor their machines and processes around the clock.
Smart Management Of Data
Making good sense of the data produced is essential for improved productivity and efficiency of manufacturing processes, and in order for this optimisation to happen, manufacturers need to be aware of which data exactly they need or want, and how to use this data to its full potential.
An easy advice to managing data smartly is to go simple, and a reference guideline is demonstrated below:
1.Determine which machines can produce what information, and from there, pick out the information that one would require
- Tip: Pick out only the most useful information
2.Using a reliable data analytics programme, determine how the programme can be used to represent the data in a way that is easy to understand
- Tip: A graph or chart would suffice, as these usually turn out to be the optimum way for people to digest information quickly
- Tip: Do not go overboard in visualising data—always keep in mind the goal for the analysis so as to not waste resources unnecessarily
3.Make an informed decision—the condensed information should be able to provide the relevant people with the required knowledge for a quick and effective decision-making process
Naturally, many more things can be done with the data for further analysis, or the data can be presented in a variety of different ways, but the main message is the same: ultimately manufacturers should be clear on what kind of data they are able to get, which kinds of data they would actually need, and how exactly they can present their data in a way they would want so as to aid in decision-making.
