How to implement Industry 4.0 within the Process Industry — by Stefan Zippel

Estimated Reading Time: 9 minutes

On June 22nd I was invited to share my insights on the very timely discussion of Industry 4.0 within the manufacturing sector. The webinar Industry 4.0: A Blueprint for Achieving a Dynamic Smart Factory was produced by Marcus Evans in partnership with Advantech and incorporated expertise from myself, Stefan Zippel, Tomas Letko (Intel) and Jim Ten Broeke (Advantech). Due to the significant amount of questions from the audience and the limited time available during the webinar, it was not possible to address them all. In an attempt to follow up on those unanswered questions, the following blog post will provide a comprehensive perspective which will hopefully clarify the challenges faced by the sector:

  • “How to create a strategy and roadmap of initiatives in 4.0 for chemical or seeds facilities?”
  • “Is there an Industry 4.0 framework for process industries like Chemical?”

A Recent View of Industry 4.0 in the Process Industry

When the notion of Industry 4.0 was first brought up at the Hannover Fair in 2013, the focus of use cases and examples were predominantly aimed at the manufacturing industry. This is not surprising when one considers that the automotive industry in Germany alone accounts for more than 800, 000 jobs.

At the beginning, the process industry perceived little benefit from 4.0, especially the entire notion of connecting production lines. The process industry, much like the chemical industry, has done this degree of process automation for more than 20 years and couldn’t understand what it was all about. In 2015 some organizations like NAMUR attempted for the first time to come up with use cases that specifically targeted the process industry. Ideas then included predictive control or even plants that could operate without a control room.

It took until 2016 for the company I worked for to start recognizing Industry 4.0 as something which should be taken seriously.

What Prompted the Process Industry to recognize Industry 4.0?

The collection of process data for the process industry is not new, however the conversion of that data to actionable information has been implemented on a very small scale, thereby creating a disturbing grave of data. With the emergence of feasible use cases and first usable products to choose from, this allowed industries like the Chemical Industry to realize the potential that Industry 4.0 could also bring to them.

Now they understand that for example the use of AI and Data Science can improve activities like predictive maintenance, energy management or process optimization. The use of mobile devices allows a broader distribution of information to allow for informed decision making on a wider scale.

Real time data is used to predict when equipment will fail, thus instead of performing maintenance on a more or less arbitrary schedule, you do it when it is necessary.

What is the Way Forward for the Process Industry?

Unlike industries such as the automotive industry, the process industry has one inherent advantage when it comes to implementing 4.0; many facilities are already collecting most of their process data either through DCS or PLC and SCADA systems and the use of MES is also not uncommon. This means that the initial investment of bringing in “IoT / IIoT” technology to perform data collection is very low or non-existing.

Many process industry facilities are already in a position to begin the 4.0 adoption process at a place that would be considered, for most manufacturers, as step 2 of the process. This enables them to gain quick value from their data by turning them into actionable information right away.

In some cases this may not even require large investments because some MES vendors are implementing new services and applications to do exactly this. Thus a simple upgrade of one’s MES system may be all that’s required.

The biggest road block may actually not be the technology you need but having the right skills within your workforce. Hiring or training data scientists, data miners and developers for dashboards will be the most difficult and time consuming challenge. Also rethinking how to conduct and understand company operations, which includes topics like cyber security, is a more urgent issue than finding some new technology.

If this can be successfully resolved, the process industry could actually take the lead in terms of Industry 4.0 convergence.

What does an Industry 4.0 framework look like for the Process Industry?

By taking all of this into account and following the adopted ISA 95 structure, the framework for the process industry could look something like this:

At the heart, lies the fully automated production facility which is and remains the ISA 95 Level 0–2. Additional “IoT/IIoT” technology could be deployed to close existing information gaps. For example, in addition to all the pumps, motors, valves and sensors already connected, there are also devices like cameras, meters for utilities (electric energy, steam, water, etc.), weather stations or geo-information from vehicles or employees that could be collected. Additionally, data from outside the production line which have an influence like raw material prices, social media, new and other mostly unstructured data could be added. This would allow for more robust and comprehensive data models and better data based information.

To transform the data into information, they will have to be transferred to the appropriate applications. The most common bridge between OT and IT for the process industry may be the MES system many have already in place, seeing as it already allows for horizontal (LIMS, WMS etc.) and vertical (ERP, PLM etc.) integration making it an existing OT-IT bridge.

Therefore we could consider an alternative idea (suggested by Joe Perino PERTEX Management and Technology Consulting LLC), one that I came across recently and find quite compelling, is to incorporate into the ISA 95 Layer 3 a so called “Digital Core”.

The Digital Core works as a central data collecting and information distribution mechanism deploying Big Data principles. It would support and integrate all the data types you are collecting including:

- Transactional and continuous processing

- Cross-disciplinary and cross-departmental work flows

- Collaboration across the plant, and between plants, with the corporate and external ecosystem

- Integrating time frames, both historical and real-time

The Digital Core would consist of multiple data engines integrated within a managed framework and usually incorporating an advanced analytics toolset. This would allow the harnessing of relational and non-relational data. It is capable of ingesting any type of big data, and runs SQL and non-SQL analytics. It also remains open to plugging in third-party solutions, whether they are a deep learning platform like IBM’s Watson or a custom program developed with Python, R or Matlab tools.

The Digital Core is like a brain connected to every data source, compelling to transform the quite static hierarchy of the current ISA 95 model to a more Cloud or Service based model, where all data sources and information users regardless of ISA95 Level are connected by the Digital Core; passing data along and receiving actionable information. And no system is necessarily higher in hierarchy than another but all have to collaborate.

The use of cloud technology either on premise, off premise or hybrid, will be a staple of any new architecture or will improve the availability and connectivity of such applications and the data transfer. The information will be accessed through mobile devices and desktop application alike based on a user’s profile. This provides better accessibility while maintaining security of the data by limiting access to the platform based on individual requirements (authorities).

People’s roles within the company will change as a result therefore equipping them with the tools to embrace change will help ensure a smooth transition through Industry 4.0.

Some specific Use Cases Examples

Energy Management

Many process industry facilities require a lot of different utilities (electricity, steam, gases, water etc.) to operate which makes them a big part of the operational costs. Energy Management has become one of the important activities to bring about a competitive edge. Currently, many just collect energy consumption, to get an ISO 50001 — Energy Management Systems — Certification. Understanding the correlation between your energy consumption and production process while factoring in environmental conditions allows for the prediction of one’s energy requirements based on your production schedule and the weather forecast. This may in turn enable the transition from long term contracts with energy providers to more shorter weekly contracts. Thereby giving you the freedom to pick and choose the best offers and reduce your cost. Another option would be to use process data together with energy data in order to optimize your production process which in turn can reduce your energy consumption.

Process Optimization and Predictive Control

This brings me to another use case, which is process optimization and real time optimization. This is not a new concept to the process industry, with some MES vendors offering real time optimization modules as well. However, it still comes down to the decision making process of whether of not there is a need to optimize.

Using AI and Machine Learning could open completely new doors in the way you run your plant. This is due to its predictive operational controls and real time optimization technology which reduce quality and safety issues, as well as energy and raw material consumption; thereby improving the operational uptime of your equipment.

Predictive Maintenance

This brings me to my third example where industry 4.0 can make a big difference, which is predictive maintenance. Similar to predictive plant control, real time data is used to predict when equipment will fail. So instead of performing maintenance on a more or less arbitrary schedule, you do it when it is necessary. This also increases the likelihood of spotting equipment failure in time which will allow for proper preparation.

Key Takeaways

There are many more examples that I could discuss, like in the engineering of a chemical plant, product development, customer service, new services and products (based on what you produce) or using Virtual reality / mobile devices to guide and assist technicians.

Nonetheless there are a few things which have to be kept in mind. Many of the use cases are relying on AI, machine learning and data science. They are very powerful and if correctly deployed will fundamentally transform how the process industry will operate. That said, there are no immediate “out-of-the-box wonder pills”. Data models need time to train before they can actually be used. They also require constant improvement and adjustment to stay relevant. This means that Industry 4.0 is not a standard project whereby you just select a solution, deploy it and then go into maintenance mode. On the contrary, the nature of Industry 4.0 dictates that you never go into a standard maintenance mode, but remain in a deployment mode.

Additionally, Industry 4.0 is not something to be used to reduce workforce; instead for it to be really effective you will need to increase your head count by hiring new and more highly specialized personnel to keep your Industry 4.0 initiative alive and well.

Therefore Industry 4.0 holds the same significance and challenges for the process industry as it does for industries like automotive or machinery. To learn more about how to start your Industry 4.0 journey or as I like to call it, your Smart Manufacturing journey, please feel free to visit

Stefan Zippel

Senior MES Coordinator

Download the webinar on demand by clicking the link below:



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store