Ever since the steam engine revolutionised manufacturing, our means of making what people need has continually evolved – from linear manufacturing to the introduction of robots and most recently to the way we use data and automation to make manufacturing smarter. Frédéric Sailly – Executive Vice President Product Management & Development, Sidel – talks about how these developments of Industry 4.0 are affecting the food, beverage, home and personal care packaging industries.
In the past, the goal of most manufacturers was to achieve high volume performance with high efficiency, paying attention to a kind of product standardisation. When change was required, switching the set-up to meet the new needs generated long periods of downtime or could even result in another piece of equipment being added to the line.
While this increased flexibility, it came at the cost of efficiency. Today’s challenge is to reverse this trend. By combining high versatility with efficiency we can help our customers meet end users’ demands for greater product differentiation. This goal can be achieved by building machine intelligence into equipment in the form of features such as prediction, aided guidance and self-adjustment. The latter one for instance lets the machine correct itself independently of human intervention when data shows that its production values are out of range, as well as when manufacturing parameters should change based on new production needs.
Built-in intelligence also means machines and lines capable of long production runs in full autonomy, with none or very limited human involvement, as well as capable to capture opportunities for late customisation, for faster time to market and mass personalisation. All in all, this creates a smoother process and a much higher level of consistency, which translates into product quality, enhanced productivity, minimal downtime and greater asset intensity.
“By combining high versatility with efficiency we can help our customers meet end users’ demands for greater product differentiation. This goal can be achieved by building machine intelligence into equipment in the form of features such as prediction, aided guidance and self-adjustment.”
Industry 4.0 means digitising the entire value chain – eventually linking the manufacturing site to the point of sale. This provides four main benefits.
First of all, the optimisation of the initial investment: for instance, by using virtual reality and simulation tools, manufacturers are reducing the time and costs involved, while anticipating any challenges the equipment might face when in action. Second, the reduction of the running costs: this is typically achieved through connectivity and built-in machine intelligence, both avoiding downtime by predicting maintenance tasks and optimising resource consumption. Third, the possibility to continuously meet demands for long-term efficiency, driven by increased product customisation and tighter time-to-market. Last but not least, maintaining a high-performance, high-flexibility line, in a world where demand volatility is extremely high.
Industry 4.0 solutions are leveraging connectivity for significantly optimised efficiency and TCO – Total Cost of Ownership. By connecting upstream to supply and downstream to logistics, it links the manufacturing site with all the steps in the value chain.
The advent of data produced by the line enables other developments, such as serialisation and traceability. This advance can be used to protect against counterfeit products and guarantee safety in liquid packaging. This is critical especially for manufacturers working with aseptic production of sensitive products.
The data gathered from the line can also be used to monitor line values, meaning scrutinising a manufacturer’s sustainability profile and identifying potential areas for improvement. How does electricity consumption compare for different production runs? Is water being used efficiently or can this be improved? This helps producers reduce losses, reach production objectives and set priorities for future business goals.
This is where digital twin comes into play. It is a virtual representation of a physical object or system. It includes also complex systems that data scientists can use to run simulations, thus predicting performances. As more complex “things” become connected with the ability to produce data, having a digital equivalent gives data scientists the chance to optimise deployments for peak efficiency and create other “what-if” scenarios. This way production time is expanded and operator’s intervention can be planned in advance. Additionally, we are investigating opportunities offered by Artificial Intelligence. For instance, based on the data as collected by the line, we can propose the best sequence of changeovers in order to increase the overall uptime.
“The advent of data produced by the line enables other developments, such as serialisation and traceability. This advance can be used to protect against counterfeit products and guarantee safety in liquid packaging.”
Through a meaningful use of data, the line knows in advance when it is short of material or if a spare part is showing signs of wear and tear, for an easier, faster and more efficient product delivery to consumers.
Maintaining performance is also possible thanks to the support of machines guiding the operators when intervention is necessary. They do that through very intuitive and simple interfaces or via a set of software and applications, autonomous root cause analysis and step-by-step guidance. The machines communicate with big-data repository in total cyber security to refine its root cause analysis capabilities, drastically reducing the downtime.