AI for process optimization

Making Smarter Decisions by Using AI for Process Optimization in Manufacturing

The use of AI for process optimization in manufacturing is gaining rapid traction. I want to believe that at this point if a company is not leveraging the capabilities tied with AI, it’s only because they are at least in the process of exploring it.

Here’s the thing, Industrial AI, which is defined as the use of advanced analytics applied to data from the production floor, to enhance manufacturing performance, is suitable for many different manufacturing industries. 

With that in mind, the idea that not all factories are using AI for process optimization at this point is mind-boggling. But I think we are on the right track. One reason for this is that I am seeing more and more “industrial influencers” making the shift. 

I’ll give you an example.

I recently stumbled across an article discussing Tesla’s move to automated factories

The Tesla Model Y, which is scheduled for production in 2020, will be part of the industrial revolution they’re undergoing: “Tesla CEO Elon Musk recently expressed his intention to make the Model Y production system a “manufacturing revolution”, Elon has been going back and forth on how much to simply copy the existing manufacturing process for the Model 3 versus trying something new, untested, and ambitious. For now at least, it sounds like Elon wants to do the latter.”. says Trent Eady

The effect that Tesla has on the automotive community is one of the largest, as they are held accountable for the catalyzation of the whole industry transitioning from gasoline to electric propulsion. To think their shift to smarter factories will have a similar effect is not only optimistic, but I want to think it’s realistic. 

This revolution that Tesla intends to undergo, is a testament to the understanding of the short and long term business benefits that come with adopting Industry 4.0 initiatives.

Not only will Tesla be optimizing processes with the use of AI, but they will be reducing costs at a great scale. 

And they’re not alone.

According to a recent report by Capgemini, by the end of 2022, automotive manufacturers expect that 24% of their plants will be smart factories and 49% of automakers have already invested more than 250 million dollars in smart manufacturing.

Sounds promising.

But still, although many companies understand the need to connect machines and gather data, it’s important to understand that using AI for process optimization is not only about gathering data. It is about aggregating and analyzing it.

This can be challenging as not all companies and manufacturers understand the need to have systems that also leverage the real-time data. But the ones who do, are definitely making faster and better data-driven decisions, which translate to optimizing processes. 

At Seebo, we are leveraging industrial AI technologies to ensure the gathered data is being put to use, primarily, for smarter decision making to optimize production. By employing digital twin analytics, automated root cause analysis, and predictive analytics - process manufacturers have visibility into production processes and their losses. What this means, is that they have the ability to not only understand why they suffer from quality and yield losses, or what these losses are but also when they will happen next. This allows them to prevent them from happening in the future. 

By being able to predict and prevent losses in production quality and yield, companies using Seebo are improving profit margins and production quality at the same time.

Ready to get started with process optimization, driven by data and machine learning?

Request a demo of Seebo Process Optimization today

Industry 4.0 automotive

The Business Benefits of Industry 4.0 in the Automotive Industry

Way back in 1913, Henry Ford, founder of Ford Motor Company, introduced the assembly line technique in mass production manufacturing. Essentially, this is what converted the automobile from being an expensive luxury, to a practical conveyance. A major revolution in the automotive industry.

Now, a century later, the automotive industry is undergoing an additional revolution, otherwise known as the fourth industrial revolution.

The fourth industrial revolution revolves around the digitization of manufacturing, and is called Industry 4.0. It is defined by the enhancement of smart systems fueled by data and machine learning. 

While Industry 4.0 has made, and continues to make, a huge impact on manufacturers across different industries, I will focus this post specifically on Industry 4.0 in the Automotive industry. 

The automotive manufacturing industry is typically divided into two sectors: car manufacturers (OEMs), and car part manufacturers (Tier 1 and Tier 2). As the cars we see on the roads continue to evolve and improve, the number of parts has grown since previous years, which naturally, has led to an increase in parts being manufactured by suppliers. 

With the complexity of today’s vehicles, and the continuous strive to perfect the end product, car manufacturers are increasingly facing quality challenges that are time-consuming and labor-intensive to resolve. 

Introducing Industry 4.0 to the automotive industry

Industry 4.0 is defined by the understanding of data captured by machines, their behavior and how to leverage that information to improve production outcomes.

And while most automotive facilities haven't yet reached the perfect state of connectivity where humans and machines seamlessly work together, the industry is beginning to embrace the principles of Industry 4.0. This is a great opportunity for them to solve quality challenges they are facing in the production line.

Fortunately, the automotive industry, is one of the more enthusiastic ones to adopt industry 4.0.

According to a recent report by Capgemini, by the end of 2022, automotive manufacturers expect that 24% of their plants will be smart factories and 49% of automakers have already invested more than 250 million dollars in smart manufacturing.

Let’s dive into the different business benefits that automotive manufacturers gain in implementing Industry 4.0 technologies:

Discovering primary causes of process inefficiencies

By implementing process-based artificial intelligence, production engineers can identify different process inefficiencies in their production line that damage quality and yield. This is done with Automated Root Cause Analysis

Automated Root Cause Analysis applies different machine learning algorithms to production line data, automatically tracing the chain of events that lead to specific production failures. This enables teams to easily investigate the causes of the failures, allowing them to mitigate the root causes.

Predicting and preventing when process inefficiencies will happen

Now, with the ability for production teams to understand the cause of specific production failures, they will want to prevent them from happening again.

This can be done with predictive analytics, which essentially translates the captured data into predictive insights. This allows for production teams to identify when specific process inefficiencies will occur, giving them the ability to prevent them before it happens. By having this ability, process teams are able to increase yield and prevent quality failures.

How Bosch Automotive Benefited from Industry 4.0

Let’s take a look at the business benefits that Bosch Automotive Diesel System factory realized after implementing Industry 4.0 to optimize production processes.

Bosch was experiencing production failures and losses, leading them to search for a way to identify bottlenecks in their production operations in order to prevent them.

As combining IIoT and big data is a big part of the digital transformation Bosch is undergoing, they connected their machinery to monitor the overall production process at the core of its plant. By using data analytics to process the data in real-time, they were able to predict production failures, enabling them to prevent future losses from happening before they even occur.

They saw more than a 10% increase in throughput, and continually improved delivery time and customer satisfaction.

The implementation of an AI-powered data solution, ultimately allowed for data-driven decision making, resulting in optimized production.


By leveraging Industry 4.0 technologies, automotive manufacturers can address processes-driven quality and throughput losses in production and assembly processes. For example, surface quality issues, coating issues, paint thickness problems, dashboard assembly issues, interiors and more, can all be mitigated. By doing so, they experience the long term business benefits that translate to increased ROI.


Citizen Data Scientist

The emerging role of the Citizen Data Scientist in manufacturing

As in many dominant fields, Artificial Intelligence has also made its way into manufacturing. With the power that industrial AI brings, manufacturing teams can leverage real-time data to optimize production processes and reduce downtime of critical assets.

Process manufacturers face rising demands for production capacity, and continually face production losses that have a direct effect on quality and yield. 

Manufacturers are increasingly turning to Industrial IoT solutions - that employ Industrial AI technologies - to quickly investigate and solve the root causes of such production losses.

When implementing a big data analytics solution on a production line, the solution will collect and analyze the data at hand. 

Naturally, as this task will entail statistical and machine learning expertise, companies must hire data scientists for the job. 

This is easier said than done.

The shortage of 250,000 data scientists in the US alone, with a staggering 29% increase in year-over-year demand for these professionals, have led to a situation where data scientists can be picky in choosing the projects and companies they’re interested in working for.

Moreover, data scientist salaries are very high, with their median base salary being $130K. It’s no surprise therefore, that LinkedIn ranks data scientists as the number 1 job in America for 2019.

The difficulty and costs in attracting and retaining data scientists has led to a new emerging role - the Citizen Data Scientist. A role often given to existing employees in an organization trained to use data analytics tools and technologies to extract insights from big data. 

What it means to be a Citizen Data Scientist in manufacturing

Gartner defines a Citizen Data Scientist as a person that creates or generates models that use advanced diagnostic analytics or predictive and prescriptive capabilities, but whose primary job function is outside the field of statistics and analytics.

Citizen Data ScientistOne of the main advantages of this role in manufacturing is that it leverages the engineering skills that already exist within the company. With manufacturing teams consisting of multiple types of engineers - such as quality engineers, process engineers, maintenance engineers, and chemical engineers - the skillset and potential are there for companies to cross-train and certify specific staff members to become their Citizen Data Scientists.

Production engineers can fill the role of the Citizen Data Scientist, bringing to bear their background in production processes and assets, math, statistics, and modeling. 

Production engineers use these skills to bring more value from analytics because they can:

  • Better understand the data and its integrity
  • Quickly assess the effect of machine learning models to a business problem
  • Identify false-positives with confidence

Simply put - the manufacturing Citizen Data Scientist can get to meaningful, accurate and actionable insights - faster than data scientists that lack a deep understanding of the production line processes and assets. 

By incorporating a  Citizen Data Scientist into your team, your engineers can now act as “power users” who may not be experts in data science, but are capable of using the tools to provide strategic and operational insights for the manufacturing business.

Citizen data scientists can essentially perform both simple and moderately sophisticated analytical tasks that would previously have required more expertise, as well as more budget.

We don’t see the Citizen Data Scientist replacing the need for data scientists. There will always be some need for “heavy lifting” AI algorithm work. 

But by filling the Citizen Data Scientist role, and deploying an Industrial AI solution, organizations can adapt their analytics models as their business needs change - ensuring their business agility - without reliance on the availability of data scientist professionals.

As the manufacturing industry’s readiness to adopt AI-powered solutions grows, we expect to see more positions of Citizen Data Scientists created and filled in most production teams. 

Production Optimization

The Role of Industrial AI in Chemical Manufacturing Digitization

One of the main ongoing concerns for manufacturers is how to enhance productivity. This includes analyzing, measuring, modeling and implementing specific actions to optimize their production lines.

The reason production optimization plays a huge role in manufacturing, is because it is a core means to increase revenue and reduce costs. Process failures in production cause companies to suffer from losses in quality and yield - which translate directly into revenue loss.

Let’s take a look at the chemical manufacturing industry, as the birth of the heavy chemical industry coincides with the beginning of the Industrial Revolution. The chemical industry comprises about 15% of the US manufacturing sector,  manufactures more than 70,000 different products, and is responsible for 90% of our everyday products.

The Challenges Chemical Manufacturers Face

Just as broad as the chemical manufacturing industry is, so are the process optimization challenges it faces.

In order for chemical manufacturers to optimize production lines, they need to address different process inefficiencies, such as the formation of undesired side products, process instabilities, losses due to impurities and more, on an ongoing basis.

Given the complexities of chemical manufacturing, it’s extremely time-consuming and difficult to understand the root causes for these process inefficiencies, let alone anticipate when they are going to happen. Often times, it is the specific behavior of the combination of multiple production parameters, or tags, that cause the inefficiency to happen.

A growing number of chemical manufacturers are turning to Industrial Artificial Intelligence solutions to identify and anticipate process inefficiencies leveraging methods of supervised and unsupervised machine learning.

According to recent research by Accenture, companies that have implemented Industrial AI in the chemical sector are seeing big benefits—a whopping 72 percent report a minimum 2x improvement in some process KPIs, and 37 percent a 5x improvement. For example, a manufacturer of Ethylene Dichloride implemented process-based Industrial AI to solve a number of process inefficiencies, and by doing so increased yield by €1.7M in less than 12 months.

With the capabilities Industrial AI has to offer, chemical manufacturers can utilize their data, improve their processes, and continually adapt them.

How Industrial AI is Revolutionizing the Chemical Manufacturing Industry

Chemical manufacturers need to identify and avoid process inefficiencies to improve chemical process control.

A production disturbance is any unintentional event in the chemical production process that leads to process inefficiencies, unplanned stoppages, rework, or scrap

By implementing Industrial AI solutions to chemical production lines, manufacturers have the ability to leverage different AI technologies that are critical to identifying production disturbances and optimizing production:

  • Real-time data connectivity and capture - manufacturers use  industrial IoT connectivity to securely connect to the production line assets and capture data in a central time-series repository
  • Process-based machine learning – manufacturers use process-based AI to get visibility into the full manufacturing process in detail, and holistically, and to discover and surface process issues that need attending.
  • Digital Twin visualization – manufacturers use a digital twin, which is a virtual representation that matches the attributes and operational metrics of a “physical” production line through the captured production-line data. This enables production teams to quickly pinpoint performance anomalies and their root cause, providing them with actionable insights, and presenting them in the context of the production line. This eliminates the need for data scientists.



ai chemical



Let’s dive a bit deeper and look at how specific Industrial AI technologies can be used to identify, anticipate and prevent chemical process inefficiencies:

The first step manufacturers should take in order to identify the specific process inefficiencies, is implementing digital twin visualization for them to easily track their main KPIs and receive actionable insights of process anomalies.  

Then, Automated Root Cause Analysis can be performed in order for them to get fast and accurate insight into process inefficiencies. The Automated Root Cause Analysis enriches historical and real-time asset data, and applies machine learning algorithms to automatically trace the causal chain of events leading to production failures.

Once process inefficiencies have been identified by using the analyzed data, it’s important to translate the data into insights. This can be done with Industrial Predictive Analytics.

Machine learning algorithms can be implemented to identify relevant events and predict their outcomes.

By having the ability to prevent specific inefficiencies and production disturbances, process teams can increase production yield while preventing failures at the same time.


By using process-based machine learning, manufacturers get focused and contextual predictive alerts. This is a huge opportunity for chemical manufacturers, since operational technology (OT) data is already well organized and captured within data historians.

Leveraging this data with process-based AI means being able to pinpoint the root cause of process disturbances with extreme accuracy, and predict process instabilities and failures before they have the chance to affect production.

So with Industrial AI, chemical manufacturers can reduce quality and production losses, saving them great amounts of time and money.


Ready to get started with process optimization, driven by data and machine learning?

Request a demo of Seebo Process Optimization today

Lean manufacturing techniques

Lean Manufacturing Techniques
in the Age of Industrial AI

Lean manufacturing is a series of principles and techniques to help improve performance in factories, production lines and machines by eliminating as much waste as possible. And more specifically, minimizing activities and products that add zero value.

The rise of Artificial Intelligence (AI) has introduced an innovative means to achieve such continuous improvements in production.

While the reward of implementing lean manufacturing techniques is great, and includes increased productivity, improved quality and minimized rework - until recently, the price has been high, and required much effort and the need for human experts.

Thankfully, as AI continues to play a huge role in leading industries, it has also entered the world of manufacturing with numerous use cases, such as predictive quality, predictive waste, and predictive maintenance.

Traditional AI vs. Process-Based Industrial AI

Before discussing how Industrial AI is helping manufacturers adopt a lean approach to manufacturing, it’s important to understand what Industrial AI is.

Both traditional AI and Process-Based Industrial AI analyze raw data generated from production lines (OT data). However, traditional AI applies unsupervised machine learning algorithms to the raw OT data, which eventually leads to a flood of alerts and many false-positives.

On the other hand, Process-Based AI, contextualizes the OT data by adding business data from IT systems into the dataset, together with the specific production process flow context, and by doing so,  builds a process-based data model.

It then applies process-based supervised machine learning algorithms, which are able to clear the noise of false-positives and pinpoint actionable insights for production teams.

The Role of Industrial AI in Lean Manufacturing

Machine learning can provide predictive insights to users - process and quality engineers - freeing up their time to focus on solving issues, rather than investigating them.

Users provide feedback to the machine learning algorithms in the form of an accuracy and relevance score, enabling the algorithms' accuracy to improve over time. This is called "Human in the loop" and leverages the engineer's production knowhow to fine tune the results of industrial AI.


Lean Manufacturing Techniques

One of the leading use cases of  Industrial AI in leading the industry to a leaner approach is by predicting waste, and by doing so - reducing it. This is sometimes referred to as Quality 4.0.

With industrial AI, process engineers can predict and prevent production waste by identifying areas of loss and prescribing focused actions that reduce product defects and inefficiencies.

This is done by employing predictive analytics and Automated Root Cause Analysis to anticipate process failures that result in wastage.

Beyond anticipating when waste will exceed thresholds, by implementing predictive simulation, process engineers test production parameters until optimal values are determined for minimizing waste and rework.

To summarize, waste is a strategic operational loss in manufacturing, but factories that harness industrial  AI technologies as an integral part of their lean methodologies can continually improve their production processes to minimize waste.


Ready to reduce waste and increase margins?

Request a demo of Seebo Predictive Waste today


Process Optimization in Manufacturing

4 Technology Pillars to Achieve Process Optimization in Manufacturing

With more and more advancements in technology, implementing an achievable process optimization plan is not as far fetched as it used to be.

The key to optimizing a manufacturing process is to embrace some of the advanced industry 4.0 technologies available today.

By understanding which technology is best for your manufacturing business, you will be one step closer to optimizing your process.

Let’s dive a bit deeper into what this means, and into the 4 main technology pillars to process optimization in manufacturing.

1. Leverage real-time data by adopting industry 4.0 technologies

The implementation of automation and use of data in manufacturing is what’s called "Industry 4.0", with use cases such as predictive maintenance and predictive quality. Industry 4.0 includes the following technologies critical to process optimization:

  • Real-time data connectivity and capture - Use industrial IoT connectivity to securely connect to the production line assets and capture data in a central time-series repository - either on-premise or on-cloud.
  • Process-based machine learning - Use process-based artificial intelligence to get visibility into the full manufacturing process in detail, and holistically, and to discover and surface process issues that need attending. By using machine learning algorithms to process and analyze real-time data, not only can process inefficiencies be identified, but they can be predicted and even avoided.
  • Digital Twin visualization - A digital twin is a virtual representation that matches the attributes and operational metrics of a “physical” production line through the captured production-line data. A digital twin of the production line enables you to quickly pinpoint performance anomalies and their root cause, providing you with actionable insights, and presenting them in the context of the production line. By having this ability, there is no need for data scientists - the system is easy-to-use and accessible for production teams.

2. Discover primary causes of process inefficiencies

As mentioned above, by implementing process-based artificial intelligence, process engineers can identify inefficiencies, such as the formation of undesired side products, process instabilities, impurities and more. This can be done with Automated Root Cause Analysis.

Request a demo Seebo Process Optimization today.

Before understanding how this will help you achieve process optimization, let’s take a look at the difference between traditional root cause analysis, and automated root cause analysis.

Firstly, traditional root cause analysis takes time - often measured in days - and expert resources from multiple teams. With massive amounts of data captured from thousands of tags every minute, it’s almost impossible to find correlations between the operational variables that lead to a process inefficiency. The longer the analysis takes - the longer the process inefficiency happens in the production line.

For this reason, production teams need a faster and more accurate way of finding early events that lead to production failures.

Automated root cause analysis enriches historical and real-time asset data, and applies machine learning algorithms to automatically trace the causal chain of events leading to production failures.

By doing so, investigation teams get fast and accurate insight into early symptoms of process inefficiencies, making it easy for them to pinpoint and mitigate the root causes.

3. Predict when process inefficiencies will happen

Having the ability to identify why process inefficiencies in your production line happen, is priceless. But if you take this one step forward, you can also anticipate exactly when they will happen.

By applying industrial predictive analytics, you have the ability to translate data into predictive insights.

Machine learning algorithms can then be implemented to identify relevant events and predict their outcomes.

For example, predicting when undesired side products will form, or when a specific process instability will happen. By doing this, process teams are able to increase yield and prevent imminent quality failures.

4. Determine the best fit process values to avoid process inefficiencies

Once we’ve understood why process inefficiencies happen and can predict them before they happen, it is fundamental to understand how to optimize the manufacturing process with these insights at hand.

Predictive simulation determines how specific inefficiencies can be avoided by simulating how processes will behave in different scenarios, and how to avoid the anticipated process inefficiency.

By using predictive simulation, process teams can:

  • Close the loop and take action on analytics recommendations
  • Adjust only the production settings that will eliminate process inefficiencies
  • Reduce the risks in mis-adjusting production settings

To summarize, the coming of age of industrial artificial intelligence, and machine learning specifically, has introduced an opportunity to harness production-line data to surface actionable insights and drive continuous improvement in manufacturing processes. And digital twin visualization makes it now possible for process engineering teams to use these insights independently of data scientists and take action in a timely manner.


Ready to get started with process optimization, driven by data and machine learning?

Get Demo


Seebo partners with Microsoft to deliver scalable
process-based machine learning solutions to manufacturers

Microsoft and Seebo, the pioneer in process-based industrial AI, announce co-sell partnership benefiting enterprise clients.

Tel Aviv, April 1st, 2019 - Seebo, a pioneer in process-based industrial AI, with solutions to predict and prevent losses in production yield and quality - today announced that it has earned IP Co-Sell Ready status with Microsoft.

Seebo is now among an elite group of global independent software vendors selected by Microsoft for intensive joint sales, support, and go-to-market initiatives.

Seebo will work directly with Microsoft’s global sales teams to increase the adoption of Seebo’s AI-based solutions among manufacturers around the globe. Specifically, Seebo will work closely with Microsoft’s enterprise sales and delivery teams to engage mid-size and large process manufacturers with the opportunity to optimize quality, yield, and throughput.

Warrick Hill, Senior Managing Director of Microsoft Europe, says “Seebo solutions deploy Microsoft Azure AI services at scale. We’re excited to be a strategic component of the digital transformation that Seebo delivers to our enterprise clients.“

“I’m excited for this partnership as it is a true testament of the strategic value we bring to manufacturers and is built on the success and collaboration that we’re already seeing in winning together.” Said Lior Akavia, Seebo CEO and Co-founder.

Manufacturers are continually challenged to overcome unexpected inefficiencies in their manufacturing processes. Process inefficiencies such as the formation of undesired side products, process instabilities, and blockages, damage production yield, quality, and availability.

Seebo predicts and prevents these inefficiencies, enabling production teams to anticipate when and why they will happen, and to understand how they can be avoided altogether. Seebo addresses common Industry 4.0 use cases such as predictive quality, waste reduction and lean manufacturing, production optimization, and predictive maintenance, with unmatched accuracy and ease-of-use.

The Microsoft IP Co-Sell program was officially unveiled at Microsoft Inspire in June 2017. Since then, it has expanded to a global scale, and includes a $250 million investment in Microsoft seller incentives, to drive engagement, growth and amplified visibility of Microsoft’s technologies.

Seebo is available on Microsoft Appsource


About Seebo

Seebo is a pioneer in process-based Industrial AI, with solutions to predict and prevent process inefficiencies that damage production yield and quality. Customers use Seebo to know:

  1. When these process inefficiencies will happen - leveraging process-based predictive analytics
  2. Why they will happen - with automatic root cause analysis
  3. How to avoid them - using predictive simulation

Manufacturers across industries – including Nestle, Grundfos, Stanley, Procter & Gamble, Hovis, Allnex, and many more – use Seebo to increase production yield while continually improving quality.



Oren Ezra

VP Marketing, Seebo


7 Wastes Plus One? The Hidden 8th Waste in Lean Manufacturing

Lean Manufacturing

The manufacturing industry is as broad as it gets, it consists of 5 different types of processes, spans dozens of verticals, and involves various methods,  philosophies, and approaches.

But all manufacturers have one common challenge they need to overcome, which is the problem of waste.

The adoption of a waste minimization mind state is what’s called lean manufacturing, and is defined as a systematic method for waste minimization.

Lean manufacturing categorizes waste to seven different categories and was originally derived from the Toyota Production Systems (TPS) in 1990. What it means is that anything that doesn’t add value (while not sacrificing productivity), is considered waste.

The 7 wastes in lean manufacturing

While lean manufacturing categorizes 7 different types of waste, it seems there is one more type of waste, “the hidden waste”, which I will discuss later in this post.

But first, let’s understand what the 7 wastes are.

The 7 waste in lean manufacturing:

  1. Transporting - moving materials from one position to another. The transport adds zero value, and there’s definitely room to minimize the costs.
  2. Waiting - one of the more serious wastes in manufacturing, simply put, waiting for goods to move or be processed.
  3. Inappropriate Processing - often, organizations use high precision equipment, which can easily be replaced by much simpler tools. And by investing in smaller, more efficient tools, this waste can be eliminated.
  4. Defects - defects eventually affect quality, which leads to loss of money, either because a product is sold for less, or not sold at all. It’s not enough to detect these afterward as this will also result in extra time and money, but rather prevent it from happening in the first place.
  5. Overproduction - this relates to when there are more produced products than demand for them. Overproduction can also lead to other wastes, such as waiting, inventory, resources and more.
  6. Unnecessary inventory - unsold products lead to extra inventory that organizations are “stuck” with, which need to be stored and take up space and transportation.
  7. Over Processing - when inappropriate techniques are used, or the wrong equipment, processes that are not required are performed, which cost time and money.

The 8th ”hidden waste” in lean manufacturing

Though the list of opportunities and potential to minimize waste in manufacturing seems comprehensive  - there is one more type of waste we see many process manufacturers dealing with, with huge potential to minimize:

  1. Process Inefficiencies - process inefficiencies are different “disturbances” in the production line that can affect quality and yield.

For example, in the chemical manufacturing industry, such process inefficiencies can be:

  • Formation of undesired side products that affect the product purity- when 2 or more reactions occur simultaneously
  • Incomplete reactions - which damage yield and quality of the finished product
  • Losses during separation - of the desired product from the reaction mixture
  • Process instability - due to blocked assets, leakages, and other asset faults
  • Losses during purification - due to the transfer of material from reaction vessels
  • And more.

7 wastes

The bad news is that these process inefficiencies come on account of meeting production goals, such as increasing product purity, preventing asset failures, increasing throughput, and much much more, but most importantly - reducing waste.

But the good news is, there is a way to minimize them. With the rise of Industrial Artificial Intelligence (AI), more and more manufacturers are leveraging process-based machine learning solutions to predict and prevent process inefficiencies.

Predicting and preventing process inefficiencies with Industrial AI

When it comes to AI, it’s important to understand the difference between traditional AI vs. process-based AI.

While traditional AI looks at raw data from production lines (OT data) and applies machine learning to it (causing many false-positives), process-based AI contextualizes the data by adding business data from IT systems into datasets -- together with the specific production process flow context -- and builds a process-based data model.

It then applies process-based machine learning algorithms, which are able to clear the noise and pinpoint actionable insights.

What this means, is that by implementing process-based machine learning - we can now understand 3 important insights:

  1. Why process inefficiencies happen
  2. When they will happen, and
  3. How to avoid them from happening again

The end result? Improved yield and quality, which leads to reduced waste.

There’s no question regarding waste being a strategic issue for manufacturers and the importance of understanding how to combat waste in the most cost-effective way.

Manufacturers, and process manufacturers in particular need to address the 8th waste of Process Inefficiencies as a core cause of waste. This requires the implementation of machine learning with its predictive analytics capabilities.

Industrial IoT Trends in 2019

8 Industrial IoT Trends in 2019

Capitalizing on technological advancements in industrial manufacturing, companies are taking bolder steps to improve growth and operational efficiency in 2019. Here’s a look at the top trends and predictions for industrial IoT in the coming year.

It’s no surprise that worldwide technology spending on the Internet of Things is forecasted to reach $1.2 trillion by 2022 (IDC).

Manufacturers are looking to solve the complex problem of consolidating all production systems – OT and IT data, BI, quality management, and production processes –  into a single data model. And they know that those who manage it successfully can beat their competitors in the process. For this reason, adoption of IoT devices and services is set to hit 20% in 2019 (IDC).

But the question isn’t “what”, it’s “how”: there are numerous Industrial IoT solutions to target a myriad of business problems, and manufacturers can only budget for a few POCs or solutions at a time.

With so many possibilities for improvement, where will manufacturers, investors, and governments choose to put their money?

Based on research, we’ve identified the trends that will continue to gain traction in 2019. These are the actions manufacturers will take to better manage operations, deliver improved products and services, and grow business smarter.

8 2019 Trends in Industry 4.0:

Going beyond POCs

If 2018 was the year of IIoT proof of concepts, 2019 will be the year manufacturers move from early proof of concepts to deploying pilots for Industry 4.0 solutions, such as predictive maintenance, digital twin, and predictive quality.

Industry 4.0 solutions are so new that we still lack data for much of the ROI from Industry 4.0 initiatives.

Here is an example:

Predictive maintenance is the headline subject at practically all of the recent and upcoming international Industry 4.0 conferences. But predictive analytics in manufacturing still requires months of collecting enough data to act upon before providing full ROI.

Furthermore, while some manufacturers have reached the predictive stage, very few early adopters have reached the prescriptive analytics phase.

But that is about to change.

With major players in food & beverage, chemicals, and other giant industries deploying Factory 4.0 solutions, there will be more information about how these solutions are effectively implemented per industry by this time next year.

The rise of industrial AI in manufacturing

AI and Industrial IoT are merging to digitize production processes to increase productivity and reduce downtime. Machine learning algorithms for manufacturing are being formulated and tailored to specific production line challenges – such as reducing production waste, improving process stability, minimizing unplanned downtime, and eliminating process disturbances

What does this mean on a practical level?

Search “AI in manufacturing”, and you’ll read articles about more accurate insights into the manufacturing process, and reaching a higher OEE than was possible through previous methods. These aren’t just superlatives – the vast scope of IoT use cases makes manufacturing the most logical field for AI applications.

Industrial IoT trends 2019

AI and machine learning are a wide umbrella term for a multitude of different algorithms and applications; many of the trends below are part of that shift towards integrating AI solutions into existing manufacturing processes.

Contextualizing OT data – and tools for contextual analysis

OT and IT have been converging for some time, and ‘collaboration’ used to be the goal – but many manufacturers are taking their operational and IT data a step further to improve the relevance and accuracy of data-driven insights.

What is that step?


The only way for manufacturers to measure the right data, and reach accurate conclusions, is by combining all the relevant operational data from the plant or line environment with business context data from the IT systems.

Are you contextualizing your OT data with the following dimensions?

  • Production process flows
  • Production recipes
  • Batch and product details
  • Quality test results

Here’s an example of data contextualization in predictive maintenance:

A food and beverage manufacturer deploys software with machine learning algorithms applied to OT data from one production line, searching for patterns that predict asset breakdown.

But this software doesn’t take into consideration alerts from quality control tests, or which batch and product is being manufactured.

So an oven could overheat for specific recipes – but without the context of the recipe the machine learning algorithm could never yield accurate, actionable insights to the production team.

In the coming year, manufacturers will be budgeting for systems that help them glean manufacturing excellence insights  through the lens of process and business data that influence the production environment.

Using digital twins

24% of companies implementing IoT solutions already use Digital Twins to increase safety and efficiency – and according to Gartner, that number is about to jump.

Digital twins are virtual copies of a physical entity that link to the entity, often in real-time. In the manufacturing world, digital twins support many industry 4.0 solutions, from automated root cause analysis, to predictive quality, predictive maintenance, inventory intelligence, and supply chain optimization.

Digital twins are most commonly used in the areas of design, modeling and simulation, so it’s not surprising that they were a buzzword in 2018, the year of Industrial IoT pilots.

In 2019, we will not only see greater adoption of the digital twin in general, but also an expansion in their popular use: more digital twins used to optimize production processes rather than the individual assets in day-to-day operations and processes.

These ‘full’ digital twins’ will incorporate process data that will help manufacturers reach more accurate insights, whether by deep-diving into individual machines or viewing the high-level process architecture to identify and address manufacturing inefficiencies.

Early adopters are already using digital twin software to improve the accuracy of predictive AI applications, and more manufacturers will adopt this approach in the coming year.

Edge computing

As devices become more powerful in 2019, more manufacturers will take advantage of local data processing and AI capabilities, also known as edge computing.

By 2020, IoT sensors and devices will generate over 5.07.5 zettabytes of data.

Manufacturers are for the most part already collecting data, but managing it via cloud computing puts a financial strain on manufacturers – not to mention the security risks of storing all your raw data in the cloud.

Edge computing helps businesses by analyzing and storing data close to its source decrease time and expenses related to data analytics, as well as improving data security.

Imagine this:

Multiple machines in one production line monitoring vibration of machine components.That’s hundreds of data points per second.

Uploading all that data to the cloud for cleansing, processing, aggregating and analysis is redundant.

In edge computing, each machine in the line is connected to an edge  computer to collect, store, and preprocess OT data.

The edge computer not only processes vibration data, for example, but also performs feature extraction to select a predefined amount and type of vibration data – highs and lows within a given time frame, for example – to go to the cloud.

Even this basic level of data analysis processing performed at the source streamlines the process of aggregating production line data to an incredible degree. There is much less historical and real-time data for machine learning algorithms to sort through, speeding up findings that could affect everything from yield to uptime to product quality.

Secure and cost-effective

Edge computing also cuts down on the cost of data storage in a cloud: a dozen data points for every ten thousand measured. Limiting the raw production data sent to the cloud also mitigates data security risks.

Mobile Industry 4.0– ERP and quality management systems.

The arrival of 5G next-gen mobile networks heralds greater adoption of IIoT applications.

Due to 5G and other advancements in mobile tech, 2019 will see a rise in real-time IIoT applications and the use of IIOT for teams once excluded from direct interaction with the technologies involved.

For example, while many ERP systems are now integrated with Industry 4.0 systems and even include applications, such as MES, most of these systems do not support all the user roles and business functions related to the manufacturing floor.

This is unfortunate, since many of the solutions forecasted to grow in adoption, such as operator productivity and inventory intelligence, must be accessible to teams on the factory floor.

Expect to see a rise in companies offering apps specializing in different user personas, such as quality management software apps, or applications with different dashboards per business role.

Supply chain optimization

What was once purely a logistical function, now has their own business models and optimization processes.

Coupled with this, Online-consumer trends have dramatically changed customer expectations for on-demand services, transparency, speed, and efficiency. Supply Chain 4.0 is a way to  meet the new demands and changing supply chain landscape through digitization.

Supply chain optimization can and does utilize many of the other  top Industry 4.0 trends for 2019: Digital twins, mobile apps, and AI-powered predictive tools. Artificial Intelligence will be imbedded in common supply chain processes.

Accurate – Digitally map the supply chain and real-time production and product data create transparency and increase the accuracy of inventory data.

Faster – Advanced forecasting tools, coupled with real-time data on spare parts supply and demand, will create process where finished products reach their destination faster.

Flexible – real-time data leaves room for flexibility in the distribution process

Securing IIoT endpoints

Enterprises are already invested in securing their OT infrastructure, to the same degree as they do their IT systems. However, the clear and present threat to organization’s cybersecurity, coupled with the boom in Industrial IoT adoption, will see OOT security and ICS security go mainstream in manufacturing plants, whatever the size or industry.

It won’t be easy; securing IoT devices or machines is becoming increasingly difficult, so much so that Microsoft recently released a list of best practices for IoT devices.

And with the proliferation of edge computing comes a plethora of new industrial IoT endpoints, i.e. devices with computational capabilities and network connectivity ( So even the bonus of securing data by sending less to the cloud comes with the added risk of increase endpoints.

Despite this, the gain to businesses in developing smart production lines is clear enough that securing data is becoming less of a deterrent to manufacturers who want to increase yield and improve operational efficiency through AI.

Forecast: Top 8 Trends in Industrial IoT in 2019 

1. Going beyond POCs

2. Rise of industrial AI in manufacturing 

3. Contextualizing OT data – and tools for contextual analysis

4. Using digital twins 

5. Edge computing

6. Mobile Industry 4.0– ERP and quality management systems.

7. Supply chain optimization

8. Securing IIoT endpoints

Industrial IoT Predictions for 2019

We’re seeing more and more deployment of Industrial IoT solutions that are revolutionizing the manufacturing landscape digitally – transforming customer relationships, differentiating offerings, and driving massive operational improvements to meet the growing demands on production.

Based on these trends, industrial IoT early adopters are positioned to be five times as likely to generate revenue from Industry 4.0 initiatives compared to late adopters. But take note – companies must first decide which business value drivers they want to contribute to. Only then, can they align their digital strategy with the business goals they are targeting in order to manage, secure, and operate IoT platforms and processes effectively.

New call-to-action