Manufacturing 4.0
Manufacturing 4.0
Manufacturing has seen many improvements over the past couple of centuries, from steam power and electricity, assembly line to the adoption of computers. All of these boil down to the two main questions “How well is/are my plant(s) running?” and “Am I making money?”. We have in the 21st century entered the age of Industry 4.0, a natural evolution from the computer-controlled machines and system, to a connected self-controlling and self-optimizing production. Transition to Industry 4.0 allows for greater flexibility, opens new opportunities, and, most importantly, gives companies a competitive advantage.
Data collection and exchange are at the basis of this evolution. Often referred to as the Internet of Things (IoT) or Industrial Internet of Things (IIoT), it is transforming manufacturing all over the world. The goal of this transition is to achieve self-controlled production and optimization of the entire value chain. It does not allow for any data islands, connecting all systems via an ecosystem of smart devices.
The emergence of the Industry 4.0 does not mean that existing plants would have to completely discard technology they have invested heavily in the past, by upgrading and integrating networking components, they can enjoy the benefits of IIoT.
OEE (Overall Equipment Effectiveness) is a framework for measuring the effectiveness and efficiency of equipment that has become standard for manufacturing. OEE is a product of Availability, Performance, and Quality. It identifies the percentage of de facto productive manufacturing time. The higher each of these indicators is, the higher is the overall OEE. It allows drilling down to the individual aspect of manufacturing and gives valuable insights for improvement. However, the reality of manufacturing these days is that OEE is often measured manually, which leads to data that is neither timely nor actionable.
IIoT allows companies not only to measure OEE in real-time but model potential improvements, without disrupting the manufacturing processes. Production flows can be visualized to allow engineering to understand points of failure and disruptions. Moreover, it allows for predictive maintenance, where machines can be serviced before they break down. Combining the widely accepted OEE with the data collected with IIoT creates a possibility for creating a significant competitive edge.
Combining networking and data collection with modern machine learning, artificial intelligence, advanced modeling, and real-time big data processing, companies can achieve innovations that seemed impossible just a few years ago.
Some of the examples of such are Predictive maintenance and holistic Product Lifecycle Management (PLM), Digital Twins.
Predictive maintenance and holistic PLM allow predictions of the right time for servicing based on the actual usage of the machines or even plants. It will enable companies to plan the timelines and conditions for repairs. By avoiding failures due to missed or late maintenance, the overall OEE increases, increasing the company’s bottom line.
The digital twin sounds simple on paper, but in reality, requires the latest innovation in streaming data processing, machine learning, and AI. The digital twin is an image of the plant/machine or even a product that allows companies to introduce improvements without disrupting production lifecycles and ensure the success of whatever the innovation companies may introduce. It could be as simple as tweaking the speed of a conveyor belt and as complicated as adding image processing for quality control.
We will consider several real-life use cases of IIoT transformations.
Ensuring project success
IIoT projects garner a great deal of interest, given the potential benefits of the outcome. However, a significant portion stall at the proof of concept phase, and very few consider the result a success. A study conducted by Cisco in 2017 shows that 60 percent of IIoT initiatives have issues in the very early stages, and only 26 percent of companies have had an IoT initiative that ended in customer satisfaction. The concern here is not the maturity or availability of the technology, but rather problems arising at every development stage, from discovery and proof of concept, through the actual implementation to the launch.
Customers typically are experts in their fields, as well as the processes and assets. They can pinpoint problem areas; however, they can’t be expected to know much about IIoT. It is imperative that at the early stages, even before creating the prototype, both the customer and the company helping them implement IIoT migration concentrated on identifying the main operational disruptors.
To ensure a successful outcome of IIoT projects involve companies experienced in the field of IIoT. Doing so would not only simplify the discovery phase, but would assure the smooth running of the transition of the customer processes to Industry 4.0, improve OEE, and would help ensure the success of the project overall.
Examples of successful IIoT transformations
Cleaning-In-Place is a process widely adopted by the food and beverage industry, as well as some chemical plants. This process ensures the cleanliness of the equipment between production cycles. This process requires machine downtime, as well as significant amounts of cleaning agent and water. Such mandatory routines are rarely subject to change, but at the same time could lead to potential improvements in terms of cost and environmentally friendliness.
By collecting and displaying the data about CIP parameters (water, chemicals, energy), operators could immediately gain insights into the process efficiency and take steps to potential optimization. Moreover, this allows modeling the most efficient way of performing CIP, which could lead to potentially reducing the time, energy, and water consumption.
By simply digitizing the CIP process, companies can benefit from significant savings and increase overall efficiency.
Paper manufacturing plants are large, complex, and highly automated systems. One of the major manufacturers was struggling with high rates of scrap. However, the complexity of the manufacturing process did not allow for an easy way to find the source of the problem. A Digital Twin of the manufacturing process was created using real-time data processing of the data received from controllers and sensors throughout the lifecycle. With the use of the most recent AI, a model of the plant was created. Based on the real-time data, as well as historical data, it was possible to isolate the areas contributing to the scrap and trace back from the faulty batches through the whole production cycle. Based on the model, potential changes were introduced in the digital twin to allow optimization without compromising the manufacturing in the process.
These two examples of IIoT projects show the range of potential integrations, from digitizing data collection to a complex AI. But both cases show the kinds of improvements companies can potentially obtain.