To the Connected Factory in Five Steps
Pave the way for Industry 4.0 in manufacturing
Lights-out factory: Not a switch but a dimmer
The vision of Industry 4.0 is a complete automation of value creation:
The fully automated production.
No human operators need to be present.
This is known as “the lights-out” factory.
The lights-out factory is in discussion since 1980. Today it is already a common practice in the manufacturing of standard products. The goal is now to drive the automation also in more complex production environments.
Many companies start out with their Industry 4.0 activities. They focus on the vision of ultimate value chain automation as a guiding star. But oftentimes they are not sure which steps they need to take towards their goal.
This post examines the steps “from industry 3.0 to industry 4.0” and the value they bring in practice.
So, reaching the lights-out factory is no matter of just flicking the switch but rather a dimmer.
Five Steps to Success
I frequently discuss with industry experts and pracitioners about Industry 4.0 implementation.
The conversation typically comes to the point where someone asks about concrete steps.
Now the easy answer is, as always: “It depends”.
Implementing smart factory concepts depend on factors like:
- The waste and challenges in production
- The products and their variants
- The production processes
But there are steps an individual approach often breaks down to. These are what I call the “five steps to success” to pave the way for Industry 4.0 in production.
Step 1: Connect
This actually comes as a shock to some, but a smart factory needs connected assets.
This is the basis to use data for automation of value creation. For best results, connect machines, peripherals and other data sources to a common network.
A common network for the assets and data bases enables them to communicate with each other. This is also known as “machine to machine communication” (M2M).
In practice, direct connection of assets goes against the long standing dogma.
PLC (Programmable Logic Controller) best practice is a hierarchical asset network. Traditional manufacturing connects assets to PLCs to the MES (Manufacturing Execution System). This has proven beneficial for control automation. But a hierarchical structure is not performant enough to handle high traffic.
But when machines connect to a common network, value comes from:
- Having a control center for all connected machines’ states
- The possibility of remote access to them
This reduces the effort to maintain an overview of the production status. Simple programming tasks or production history checks become easier as well.
But only connecting the assets makes solely use of the current state information.
Step 2: Collect
After after achieving connectivity comes the collection and storage of production data.
Already common practice in the industry, it enables traceability. So production and quality data can be archived per product. This is especially relevant in safety and mission critical products.
Several options for data collection can be used:
- In a local buffer on the individual machines (common practice).
- On an on-premise database server. This is favourite of cautious customers, who want to be sure that no data leaves the plant.
- In a global cloud infrastructure. This is something which is slowly gaining support also in traditional industries.
A smart combination of data storage achieves data redundancy. This becomes imporant to reduce the risk of data losses. Optimizing storage can also reduce network load and total storage requirements.
The security of data storage and transport is important to clarify. Especially when handling sensitive and confidential data.
The results of each performed task can contribute to the product’s digital shadow.
An example is the documentation of assembly steps in automotive. Their results are archived together with the identifier, the vehicle identification number.
Step 3: Visualize
Industry 4.0 applications rely on data. But they not only aim to store it but also to generate value.
A first use for data is its visualization. It is common practice to display data on pre-defined dashboards.
They allow the user to “slice and dice” the data. Meaning, filters and sorting help them to reach first insights, like:
- The most common errors per machine and per shift
- The total production quality
- The most common quality diminishing machine errors
But in the end, the value these dashboards bring to production depend on the user. Because the software tool does not interpret the data, the human expert needs to.
This requires deep process knowledge to reach the right conclusions and improve production.
Step 4: Analyze
What if software would contain all the expert knowledge? — Then it could support the operators with immediate actionable information.
This is the goal of step 4 — which is also where most of the industry is currently working on.
It’s all about combining the available asset data with process knowledge. So, emulating how an expert would use data to generate insights.
The main benefit for the manufacturing is immediate and around-the-clock access to insights. This enhances uptime, productivity and production quality. Optimization tasks can now be carried out by regular skilled employees.
An example of using data to generate insights is “predictive maintenance”. This schedules maintenance tasks based on the machines condition, derived from data.
Step 5: Automate
As a last step, equipment is capable to optimize and maintain itself.
In this step we see one of Industry 4.0’s missions come to life. The factory where machines support all organization and decision making.
Line operators would only sign off on tasks such as parameter changes. The machine would then carry them out by itself.
Still, a maintenance crew takes care of the assets. But they are coordinated by machines requesting service.
Strategic implications to equipment manufacturers
With analytics and enhanced automation we will bring intelligence to software. But it is crucial to note that the required logic stems from a company’s intellectual property. It makes use of experienced employees and product data.
This is the reason why many data companies fail to deliver value on their own. Because understanding of the practical processes is key.
Equipment manufacturers currently see a strategic benefit from this situation: They are able to position themselves as competence suppliers.
Instead on a hardware focus, they can now excel in building, guarding and monetizing algorithms.
Daniel Sontag connects the bots: As Industry 4.0 lead and manager for connected products, he does what he loves — tying business to tech, and theory to practice.
Hi, great you enjoyed the article! Feel free to give the applause button a few good clicks or leave a short response below, thanks.
Stay tuned: On The Industry 4.0 Blog and on LinkedIn
