Augury’s Process Health Solution, Process Navigator, is steering itself to market after a series of successful proof-of-concepts – including one that translates into saving millions annually on a single separation column. “We are in the middle of an exciting revolution in how manufacturing uses real AI,” writes Iftah Levy, Augury’s General Manager for Process Health. “We’re quickly improving how we can help operators make reliable real-time decisions that align with overall business goals.”

With Process Health, we aim to help Operation teams stabilize and optimize core manufacturing processes – in the name of maximizing yield and minimizing waste and energy use. This is our way of helping manufacturing become significantly more efficient and sustainable.

Next-Level APC

Process Navigator is the platform we built to do this. As a real-time solution using cutting-edge real AI, it provides operators with ever-evolving insights to help them attain and improve their often-conflicting objectives.

Watch the product webinar on Process Navigator:
‘Every Objective, Every Shift, Every Time: Unlock Process Health’

In many ways, Process Navigator can be seen as a new generation of Advanced Process Control (APC), which is a rather broad range of tools and techniques that began to be developed in the 1980s to address opportunities to improve performance and reduce costs. However, the differences are also profound. Process Navigator enables much quicker deployment and increased flexibility. It also supports non-linear optimization, and can not only learn but also be the foundation for full automation down the road.   

“You can find the sweet spot that optimizes both yield and sustainability goals.”

From Paper Maps To Full Automation

We often use automobile navigation as an analogy to describe what Process Navigator does. Remember the good old days of charting out your vacation routes on a paper map? Remember when GPS was introduced? Both approaches were, of course, super handy but also super static. We humans still had to do a lot of heavy lifting. 

Life got a lot easier with real-time turn-by-turn navigation, as we now enjoy using Google Maps, Tom-Tom, and all the rest. And that’s what Process Navigator represents for manufacturing: your best route is automatically charted out in response to real-time conditions. 

Of course, you can always choose not to take a turn. The system will not chastise you but merely recalculate your new best future scenario. And with manufacturing, instead of taking the scenic route, you can find the sweet spot that optimizes both yield and sustainability goals. 

And as with automobiles, the next step is fully autonomous driving.  

A Great Day For Maximizing Yield

A recent breakthrough with one of our early adopter customers, a major global oil refinery, is an exciting example of Process Health’s overall potential.  

As a proof-of-concept, Augury’s Process Navigator was deployed in a single separation column to divide propane-propylene into its component units. Our system offered a recommendation every two hours over a single shift, with three of the four recommendations implemented by control room operators during that time.

The result: purity levels remained high and within specs, while yield increased by three tons per hour – an increase of a whopping 9%. Taken on an annual basis, this translates into millions of dollars of additional capacity and savings for each separation column.

“The yield was so high that the collection tanks filled up quicker than expected, and the flow lines had to be shut down to avoid catastrophe due to excess success.”

If It Works, Scale

In fact, the yield was so high that the collection tanks filled up quicker than expected, and the flow lines had to be shut down to avoid catastrophe due to excess success. The company’s CEO was immediately won over: “Let’s also use it for the night shifts. And let’s start discussing how we can move to direct control.”

Now, as we expand our lines at the facility and switch to 24/7 use of our platform, we hope to share even more impressive numbers in the short term. However, a key challenge is already being overcome: the team is adapting their work processes to increasingly use Process Navigator to optimize their line. 

Larger Challenge: Building The Trust

Just as with car drivers, it will be a while before machine operators are all willing to hand over their controls entirely. Sure, those in the C-suite who worry about a diminishing workforce – in terms of both size and experience – would perhaps love to fast-track to full automation. But that’s not realistic.

Watch the product webinar on Process Navigator:
‘Every Objective, Every Shift, Every Time: Unlock Process Health’

What is realistic, albeit still challenging, is using technology to align the company’s goals more with the operators. Today, operators don’t necessarily make decisions based on what resonates most with company objectives. They want stability. They want to finish their shift without too many hiccups. They don’t always want to optimize performance or balance conflicting and often ever-changing objectives.  

“And we do all this with a simple and straightforward interface.”

Heavy Lifting Made Easy

In any case, there’s too much going on in terms of data for a single human to track, evaluate, and decide on the best course of action. And our system does this all for them. After establishing the set points around production priorities, our system can calculate the best route and adjust that recommendation based on any changes.

And we do all this with a simple and straightforward interface (also a relative rarity in the world of manufacturing in general, and APC in particular.)  

AI As Plant’s Go-To Guy (One Who Never Retires)

As we continue to test and develop our system with other early adopter customers from a range of industries, we are not only fine-tuning the tech but also improving the required change management to help operators better appreciate Process Navigator as that trusted go-to-guy who makes their day easier and more fulfilling while also aligning with the larger business goals.

So, yes, we can safely conclude: we’re excited. 

Reach out if you’re also excited.
Or start with watching the webinar: ‘Every Objective, Every Shift, Every Time: Unlock Process Health’.

The post Process Navigator: AI As Co-Pilot – And Your Plant’s Go-To Guy  appeared first on Augury.

As an innovation and digital manager at the biggest dairy enterprise in Israel, Talli Zahavi brings big-data smarts to manufacturing dietary staples consumed daily by millions of people. “When I bring innovation to production, the production line creates more for our customers.”

Talli Zahavi is a big-picture thinker who knows that solving real-world problems requires getting down into the details.

After earning her degree in Information Systems Engineering from the prestigious Technion – Israel Institute of Technology, Talli first worked for Amdocs, a multinational telecommunications technology company. After about a decade with the company, she returned to the Technion to pursue her master’s degree, which she earned in 2010 with honors, focusing on strategic management and entrepreneurship. Then, she joined the Technion as a teaching assistant, and also held multiple roles in their Knowledge Center for Innovation.

During that time, Talli developed a passion for the value that innovation and technology can deliver all companies and economies. “Israel is well known for our high-tech sector,” she says. “And it’s undoubtedly important – it’s the engine of our economy. However, we cannot rely on just one industry to drive our economy.”

“We dive into the challenges that don’t have solutions and scout out new technologies that may bring value and a competitive edge to our business.”

As the old adage goes, “If you aren’t growing, you’re dying.” Talli appreciates that in today’s world this is particularly true. The lessons she gained at the Technion and the Knowledge Center for Innovation helped her evaluate what industries and sectors were ripe for innovation. Manufacturing stood out. “If industrial manufacturing companies are not strong enough, then the economy will not survive. So I began learning about manufacturing. And for Israeli manufacturing to be competitive, we’ve got to innovate.

When a project manager role leading digital innovation opened up at Tnuva, Israel’s largest dairy manufacturer, Talli saw an opportunity she couldn’t pass up.

Leaning Into Manufacturing Challenges

“The work our team does is so exciting,” Talli says. “We dive into the challenges that don’t have solutions and scout out new technologies that may bring value and a competitive edge to our business. This is how we came across Augury’s Process Health solution. Our dairy technologists were looking for ways to increase the yield and quality on some of our production processes. We evaluated different potential solutions and narrowed it down to Augury.”

Talli’s team deployed Augury’s Process Health solution at their Alon Tavor facility on one of their largest cheese production processes. Process Health is designed to analyze process performance and provide insights on how to improve it. This includes guidance on adjusting things like temperature, duration, and adapting to raw material properties. 

The deployment with Augury was successful. “Now it is used regularly to identify those small changes we can make to optimize our processes and output – something we couldn’t do previously.” Tnuva is now looking to extend the work they’re doing with Augury and implement Machine Health technology in their sites, too. 

The value Talli and her colleagues bring to Tnuva extends beyond the plant floor. “When I bring innovation to production, the production line creates more for our customers. In the case of Tnuva that means more milk, cottage cheese, mozzarella, brie, camembert, goat cheese, and even many lactose-free products to each Israeli household at every meal.”  

“In IT, we use and talk about big data, AI, and machine learning all the time. But we are a manufacturing company, and our employees don’t necessarily know what these terms mean and how they can be helpful in production. So we created a training program, and people throughout Tnuva could apply to become a part of it.”

Spreading the Innovative Mindset

Talli is passionate about creating a mindset shift around innovation not just for Tnuva’s IT Teams, but for the entire company. Developing “citizen data scientists” is an example of this.  

“In IT, we use and talk about big data, AI, and machine learning all the time. But we are a manufacturing company, and our employees don’t necessarily know what these terms mean and how they can be helpful in production. So we created a training program, and people throughout Tnuva could apply to become a part of it.”

This was no ordinary training program. The employees who were selected to participate attended courses about machine learning and then created a project applying what they learned to real life environments. Talli supported the process, ensuring participants had the tech infrastructure needed for their project.

“My job was to introduce these students to new technology, escort and support them as they evaluated different challenges, and mentor them as they experimented using new technologies to solve those challenges.”  

The best part? The training program included an element of competition. Employees presented projects to Tnuva’s management, and prizes were awarded for their innovations.

“I think in the 21st century, data science has to extend beyond the IT department,” Talli says. “People throughout the organization need to know the added value, strength, and benefits it can bring to their jobs.”

The post Bringing AI Manufacturing Innovation to the Table with Talli Zahavi from Tnuva appeared first on Augury.

For Augury, partnering with Azure was more than an ecosystem move to make it easier for potential customers to adopt our Machine Health and Process Health Solutions. It also provides the tools and the means to get closer to our customers so we can better innovate solutions for their benefit, according to Augury’s General Manager for Process Health, Iftah Levy.

Azure Makes Business Sense

The business case for Microsoft Azure is well-established. Basically, it won the race against Amazon and Google when it comes to ERPs. It’s now home to the databases of the vast majority of industrial users and provides a compelling selection of tools to simplify processes within your business.

We partnered with Azure for the same reasons many companies chose Azure. It’s easy to buy. And it’s easy to scale as you grow. Capacity is no longer an issue since Azure can scale with you. Plus, if you are already a client, you can use their rather ingenious credit system to rapidly adopt other solutions without necessarily paying more money. And by providing a one-stop shop for solutions, it works to simplify your ecosystem – which is often a bonus from a cloud perspective. 

In short, Azure lowers the barrier for potential manufacturing clients to take advantage of Augury’s Machine Health Solution – advantages recently outlined in the article ‘How Augury’s AI-Powered IoT Monitoring Solution in Azure Marketplace Can Deliver 3x-10x ROI’.

But there’s more… 

The Goal of Process Health: Improving Yield And Quality Without Increasing Energy Use (Everyone Wins)

Azure also makes innovation sense. Augury will be expanding its offering with new Process Health technology. And I lead a lean and cross-functional team – essentially a startup within a startup (if you can still call Augury a startup) – to make this happen. 

We are currently enjoying a breakthrough around Process Health with an early adopter client. And we will soon be introducing an exciting new technology to the market that will help Operation teams stabilize and optimize core manufacturing processes. By helping maximize yield and minimize waste and energy use, our solution will contribute to manufacturers becoming significantly more efficient and sustainable. So, stay tuned… 

How Azure Is Saving Us A Lot Of Engineering And R&D Time

And in our quest for Product Market Fit, we have been able to leverage Azure to streamline and advance our R&D much quicker than in any other scenario. We use various Azure tools – such as Integration runtime in Azure Data Factory – to access and transfer client data. More recently, we’ve been applying Azure MLOps to orchestrate all the cutting-edge AI we use now. 

For us, it goes beyond enjoying a simplified ecosystem that can take on the full life cycle of the data and the AI. To run an experiment in the past, it would take a week just to prepare the data. Now, using Azure ML Ops, the process is fast and seamless. We’re saving a lot of engineering and R&D time by using these services. 

Sure, we have to pay for it, but Azure is a real enabler for us to move fast to market (and without breaking too many things). 

In other words, we’re that much closer to using less energy. 

Read more about Process Health.
Or visit us at the Azure Marketplace.

The post Azure Not Only Makes Business Sense. It Makes Innovation Sense appeared first on Augury.

In a recent Plant Services feature, Augury’s VP of Strategy Artem Kroupenev is interviewed about the current AI applications for manufacturing. While machine health continues to lead the way in terms of rapid scalability, process health seems destined to catch up quickly.

“In the world of industrial technology, we’ve all heard that software is eating the world (famously predicted by Marc Andreessen in 2011). But this takeover by software and associated service-based business models plus changes in how software is developed has laid a perfect foundation for artificial intelligence (AI). So will AI soon be eating the world?” writes Anna Townshend in her Plant Services feature ‘Machine Health And Process Optimization Applications Take AI By The Hand’.

To find an answer on how AI is transforming manufacturing she interviewed Artem Kroupenev, Augury’s Vice President of Strategy. He broke down the current scenario into two different main applications – machine health and process health – with machine health (via predictive maintenance) leading the way.

The Economies Of Scale

“We have found that within the machine health space, especially around rotating equipment, after having built enough of a library of different failure modes on different machines, we can see that there are a lot of similarities that actually benefit from economies of scale and having a very large database,” he adds. In other words, a pump tends to fail in the same way independent of the factory or industry it’s being used in. 

However, process health remains trickier since it’s much more specific to each production line. “Even identical processes and identical production lines could have different objectives,” says Artem. 

For this reason, machine health has generally proven to be better at scaling faster. But Augury is currently working hard in making process health technologies more rapidly scaleable. “There are many common denominators across different types of processes, but the way you build the AI solution for process engineering needs to take into account the differences and be flexible enough to be able to provide value across a number of different processes,” says Artem. 

Every Case Is Different

However, there are still cases when machine health applications can lag – such as when more established manufacturers have to work around with their legacy systems. “A machine health application actually moves quicker in some cases, for example, in the food and beverage industry versus something like oil and gas, where they have 30 years of reliability practices,” says Artem. “There’s a little bit of change management that needs to happen.”

Conversely, a more established company may already have the right data collection and measuring protocols in place, which can allow them to move faster with a process health solution. 

Meanwhile, there are more than enough success stories out there to inspire manufacturers to more actively chase full digital transformation. Look at Dupont. This company has embraced AI-driven machine health, albeit backed by human vibration analysts, to now enjoy equipment predictions that have proven to be 100% accurate.

Who wouldn’t scale in such a scenario?

Read the full article: ‘Machine Health And Process Optimization Applications Take AI By The Hand’.

The post AI Is Already Eating The World Of Manufacturing  appeared first on Augury.

Food manufacturers can now use Industry 4.0 technologies such as Predictive Quality to anticipate and proactively address problems before they arise – boosting yield, efficiency and revenue in the process.

The Case For Predictive Quality

Food quality has always been a critical factor in the food production process. Food manufacturers are required to abide by stringent quality regulations, inspections and statistical quality control methods. 

The impact of poor food quality is severe, with direct bottom-line consequences. Poor product quality reduces production yield, can damage the company’s brand and reputation. More immediately, quality issues are one of the most common causes of production losses for food manufacturers, together with other, largely process-driven losses such as waste. These losses translate into a significant dent in yearly revenues.

Conversely, meeting high-quality standards can reduce manufacturing costs – internally and externally. Internal costs emerge from problems associated with the product before it is delivered (e.g., shortages, waste, and delays). External costs arise post-delivery – through recalls, lawsuits and warranty costs – and constitute a major expense for the food industry, resulting in a staggering $7 billion loss annually.

Food manufacturing operations turn to processes and tools to sustain high-quality production in addressing regulatory compliance, risk prevention, and product traceability. While tools such as statistical quality control (SPC) have been used since the early 1920s to identify issues as early in the manufacturing process as possible, they deal with problems that have already occurred. 

Predictive Quality is an emerging category of Industrial Artificial Intelligence solutions, that provide manufacturers with the means to significantly reduce process-driven losses in quality and waste, by pinpointing the root cause quickly and with a high degree of confidence, and preventing those losses before they next occur. 

Defining The “Quality” In “Predictive Quality”

In the food industry in particular, “quality” can mean a number of things. For example, there are laws governing food quality in many countries. There are also international regulations which are important regarding globalization and the increasingly complex food supply chains. These laws and regulations assure food safety and a minimum level of quality for the health and overall benefit of consumers.

While certainly a very important topic, this is not the quality challenge that Predictive Quality addresses. Rather, Predictive Quality tackles production losses caused by inefficiencies in the process itself (rather than due to individual assets, for example).

In fact, Predictive Quality isn’t limited to quality-related losses, but also addresses other common process-based production losses. In the case of the food industry that typically includes waste and yield. In other industries, Predictive Quality addresses anything from throughput, to emissions levels to energy consumption. The common thread is that these are all losses which – like the consistent quality and waste losses experienced in food manufacturing – are caused by process inefficiencies.

Industry 4.0 and Predictive Quality To The Rescue

Fortunately, food factories can overcome many of these quality control hurdles with the use of Industry 4.0 technology, specifically predictive quality. 

Predictive quality technology provides manufacturers with 3 key capabilities to optimize their processes, thereby minimizing losses due to quality issues, as well as other process-driven losses (waste, yield, etcetera).

• Automated root cause analysis
  Predictive Quality provides the process engineer with the tools to reveal previously unknown root causes of quality issues in the production line, via automated root cause analysis. Most food production lines have certain recurring losses that can’t seem to be traced to any specific visible cause. This is particularly true for more complex food production processes. That’s because with so much complex data to analyze, process engineers and experts are largely left to follow their intuition in selecting which of the hundreds, even thousands, of data tags to investigate.
  
  Naturally there will always be a blind spot – and the more complex the inefficiency, the less likely even the most eagle-eyed process expert is to find it. By contrast, a predictive quality solution uses Artificial Intelligence to do what human beings can’t: continuously analyze all the relevant data (ERP, MES, Quality systems, data lakes, etcetera) at scale – including complex interrelationships between the different data tags. This provides process experts with the ability to pinpoint the precise combination of factors that are causing a particular quality problem, no matter how complex.

• Predictive recommendations
  A key component of Predictive Quality is predictive recommendations. Predictive recommendations use the insights from root cause analysis to identify the optimal process setting. This will enable manufacturing teams to minimize quality issues as much as possible.
  
  Through continuous, multivariate analysis of production data, predictive recommendations will provide the precise optimal range of values for any given combination of tags.
  
• Proactive Alerts For Real-Time Action
  Of course, all of this intelligence must be translated into timely action to provide real value. Proactive alerts are delivered directly to production teams in real-time, as soon as a process inefficiency emerges – i.e. a problematic combination of tag behaviors, as identified by the predictive recommendations. This enables front-line manufacturing staff to act to prevent losses before they occur. Ideally, the alert should include as much actionable information – for example, not only identifying the problem, but also providing clear Standard Operating Procedures for addressing them.

Predictive Quality empowers quality teams to anticipate and proactively address quality problems before they arise – making sense of complex data patterns to determine areas of greatest quality risk and assign production floor resources before risk becomes reality. 

The business gains of Predictive Quality are clear and compelling, providing food manufacturers with a competitive edge in an era where every last drop of efficiency counts.

Want to learn more about how food manufacturers are succeeding in drastically reducing quality and waste losses? Reach out.

The post Predictive Quality: Why is it Critical for the Food Production Industry? appeared first on Augury.

With production optimization as the key route to increasing revenue and reducing costs, chemical manufacturers are increasingly turning to AI-driven solutions to keep assets running and streamline process.

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.

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.

Proven Success Story

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.

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.

How to Use AI for Predictive & Prescriptive Process Optimization

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

• Implement Digital Twin Visualization

The first step manufacturers should take to identify specific process inefficiencies is implementing digital twin visualization. This allows them to easily track their main KPIs and receive actionable insights into process anomalies.

• Perform Automated Root Cause Analysis

Automated Root Cause Analysis can then be performed to gain fast and accurate insights into process inefficiencies. This approach enriches historical and real-time asset data and applies machine learning algorithms to automatically trace the causal chain of events leading to production failures.

• Translate Data into Insights with Industrial Predictive Analytics

Once process inefficiencies have been identified using the analyzed data, it’s important to translate this data into actionable insights. Industrial Predictive Analytics can achieve this.

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.

Bottom Line: Save Time And Money

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? Reach out.

The post The Role of Industrial AI Solutions in Chemical Manufacturing Digitization appeared first on Augury.

With more and more advancements in technology, implementing an achievable process optimization plan is no longer farfetched. However, you first need to find the right technologies and approaches that work best for your particular manufacturing operation.

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 four 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.

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 Optimization is Required

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? Contact us!

The post 4 Technology Pillars to Achieve Process Optimization in Manufacturing appeared first on Augury.

With new AI-driven Process Health technologies, steel manufacturers can now achieve the seemingly impossible: increasing throughput and yield, while cutting CO2 emissions.

By having a greater understanding of the production process, a steel manufacturer was able to reduce blast furnace emissions by 3.5% and increase blast furnace productivity by 2%. As a result, they enjoyed €1.76 million savings on a single line while reducing energy intensity by 1.5%.

These are numbers that make sense to everyone. How did they do it?

The Challenge Of Conflicting KPIs

Like many steel manufacturers, this factory was in a constant race to throughput, together with other key objectives like energy efficiency and yield.

While their team had optimized the process considerably, the complex, dynamic nature of the production process meant that throughput, yield and energy levels were still unstable. And of course, such a dynamic process produced very messy data – further complicating matters for the process experts.

What’s more, recent ambitious targets to reduce CO2 emissions added yet another layer of complexity.

This challenge was particularly stark at their blast furnace, where process experts struggled with maintaining blast furnace productivity while decreasing coke rates and cutting emissions.

Their question became: ‘How can we increase throughput while decreasing emissions?’ And they were fully aware that these two objectives appear to directly conflict with each other.

Their answer: calling in an AI-driven Process Health solution.

Unlocking The Full Potential Of Their Production Line

After an initial meeting with the Process Health team, the decision was made to start at the blast furnace, as this is where the most immediately-addressable emissions and throughput losses were identified. 

After connecting to the production line data, the team could create a digital model of the entire production process, which in turn allowed algorithms to understand the intricacies of the blast furnace process, and in doing so provide accurate insights from their data.

Based on these insights, a set of clear, actionable recommendations could be offered to help achieve both main goals: increase blast furnace productivity (throughput) and reduce emissions.

Let’s break it down: 

1) A Single Metric for Global Efficiency

First off, one has to create a multidimensional objective model: a unified metric for overall efficiency at the production line.

This metric takes into account all of the factory’s production objectives, as well as any other necessary constraints – from blast furnace productivity to emissions reduction, energy intensity, and coke rate (yield).

2) Quantifying Untapped Potential at The Line

Using the multidimensional objective model, the process experts could finally identify precisely when their production process was operating more or less efficiently than average.

Specifically, they found that for 38% of the time their line was performing above their average across all objectives – including both high throughput and low emissions!

It was clear, then, that the potential for improvement existed. They just needed to replicate the conditions that led to those higher efficiency levels.

3) Identifying The Most Important Process Parameters – And Their Optimal Ranges

Next, one needed to calculate an Operating Envelope, which detailed the precise process ranges and set points that would optimize all their objectives – resulting in higher blast furnace throughput, as well as lower emissions and optimal energy efficiency.

Meanwhile, another keen insight was revealed to the manufacturing team. The blast furnace was already achieving that more efficient Operating Envelope 27% of the time! This meant that the target was even more realistic than they had imagined since their production line was clearly capable of it.

All that remained now was to ensure that the process remained within the envelope more often. Having revealed the full, hidden potential of their production line, the process experts could now give clear instructions and recommendations to the operators. 

4) Preventing Inefficiencies Before They Happen

To ensure these ideal conditions are maintained on the line, the process experts then created Proactive Alerts, which alert the production team to any inefficiencies as they occur. These alerts tell the team precisely what tags need adjusting, and also include Standard Operating Procedures – so operators know precisely how, where and when to act to prevent inefficiencies and continuously maintain the optimal process settings.

5) Onward And Upward With Continuous improvement

Of course, the key to Industrial Artificial Intelligence is a focus on continuous improvement – not a one-time benefit. Once implemented on the line, an AI-driven Process Health solution can continuously monitor the process and adapts to any changes in the line.

In addition, the process experts can also apply an Impact Analysis tool to monitor how changes on the line impact its performance over time. This enables them to refine their existing Proactive Alerts and add new ones, as well as to adjust their production objectives and constraints as appropriate.

Result: Global Process Efficiency

Armed with the optimal set points, and the ability to physically maintain those set points on the line, the manufacturing team were able to increase blast furnace productivity (throughput) by 2%, while lowering energy intensity by 1.5% and maintaining a stable coke rate.   

This resulted in €1.76 million in savings and extra profit on that single production line.

At the same time, they also reduced annual emissions by 3.5%!

In addition, by knowing the Operating Envelope, their process experts gained a deeper understanding of their production processes in general, with concrete metrics and recommendations. This saves time and effort, as process experts no longer have to spend hours theorizing and guessing the root causes of process inefficiencies.

Are you interested in learning more about unlocking the full potential of your steel factory? Contact us today!

The post Mission Impossible? Increasing Blast Furnace Productivity While Reducing Emissions appeared first on Augury.

Developed to maximize production efficiency, Lean Manufacturing meticulously broke down waste into different categories – to better confront this enemy. Today, Process Health Solutions can look at how manufacturing processes can impact each of these types of waste – without sacrificing your other KPIs in the process.

Leaning Into Waste Minimization

The manufacturing industry is as broad as it gets. It consists of five different types of processes, spans dozens of verticals, and involves various methods, philosophies and approaches. But all manufacturers have one common challenge: the problem of waste. 

And that’s what Lean Manufacturing is all about. In fact, it’s defined as a systematic method for waste minimization. Originally derived from the Toyota Production Systems (TPS) in 1990, Lean Manufacturing considers everything that doesn’t add value as waste.

The 8 Wastes, According To Lean Manufacturing

Originally, Lean Manufacturing categorized waste into seven different categories. But later, many added an eighth category: the waste of wasted human potential.

1)     Transporting: Moving materials from one position to another adds zero value. So, avoid it. 

2)     Waiting: Boring but true… Waiting for goods to move or be processed is a big waste of time. 

3)     Inappropriate Processing: Often organizations use high precision equipment in circumstances where much simpler tools can do the job as good or better. Keep it simple!

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. And since detecting these defects afterwards is too late, it all comes down to: Prevention!

5)     Overproduction: The old law of supply and demand… Why produce more goods than people want? Overproduction can also lead to other wastes, such as waiting, inventory, resources, etc. 

6)     Unnecessary Inventory: Unsold products result in extra inventory that organizations are “stuck” with – taking up space and/or transportation. 

7)     Over Processing: When inappropriate techniques and/or equipment are used, unnecessary processes are performed – which still costs time and money.

8)   Human potential: Human talent and ingenuity is a beautiful thing. Use it. The training and empowerment of frontline workers works! They are in the best position to both identify and solve problems.

In Search Of The 9th Waste: Process Inefficiencies

Though the above list of opportunities and potential to minimize waste in manufacturing seems comprehensive, there’s an additional type of waste many process manufacturers deal with: process inefficiencies.

Process inefficiencies are different “disturbances” in the production line that can affect quality and yield. For example, for the chemical manufacturing industry, such inefficiencies include:

• Formation of undesired side products that affect the product purity (such as when two or more reactions occur simultaneously)
• Incomplete reactions that damage yield and quality of the finished product
• Losses during separation of the desired product from a 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
• Etcetera

The bad news is that these process inefficiencies are often caused by the pressure of meeting production goals, such as increasing product purity, preventing asset failures, increasing throughput, and – most importantly – reducing waste.

But the good news is manufacturers can now leverage AI-driven Process Health Solutions to predict and prevent these process inefficiencies. Hence, you are enabled to be more strategic when it comes to your production lines in minimizing waste without killing other KPIs.  

From Data Chaos to Actionable Insights

When it comes to AI, it’s important to understand the difference between traditional AI versus 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 Health 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 a Process Health Solution, we can now understand three important insights:  

1)     Why process inefficiencies happen 

2)     When they will happen, and

3)     How to avoid them from happening again  

Armed with this big picture of the production line, manufacturers can now find the best and most balanced way to reach multiple objectives – including minimizing waste. 

In other words, Lean Manufacturing just got a whole lot leaner.

If you are interested in learning more on how to increase yield while decreasing waste, please reach out.

The post The Art Of Minimizing Waste While Maximizing Production appeared first on Augury.

Quality and yield losses are a perennial problem for most chemical manufacturers. Some just see this waste as the price of doing business. Others, like the global manufacturer in this case study, used customized AI to prevent these losses from happening in the first place – saving one million euros annually in the process.

For chemical manufacturers, production losses can come in many forms: quality variabilities, impurities, incomplete reactions, losses during separation and purification, etcetera, etcetera. These production losses are costly, hurting manufacturers’ bottom line and sucking up precious time and resources from their manufacturing teams.

But what if you could prevent these losses from occurring in the first place? That’s what one global manufacturer of Ethylene Dichloride managed to do — saving nearly one million euros each year on a production line that, until then, had been suffering significant annual losses.

The primary problem was a toxic side product (trichloroethane) that kept forming during the production process. Despite the best efforts of the company’s process experts, the cause of this process inefficiency remained frustratingly elusive for years.

Generic AI and Analytics Tools Make No Sense

Part of the problem was that the production process in question was relatively complex, with some 4,400 data tags. Advanced as they were, the analytics tools used by their process experts still limited them to conducting ad hoc analyses of a select number of tags. This was of little help, since once their existing theories were exhausted, they had no idea where to continue looking for the root cause of the problem!

Due to the significant financial impact of this inefficiency, the company decided to invest in Industrial Artificial Intelligence. But attempts to use generic AI solutions still failed to uncover useful or accurate results, due to the complexity of their production process. These solutions were simply not made for continuous manufacturing, and were therefore unable to cope with the unique complexities of the process and resulting data.

As their VP Manufacturing noted, “These vendors touted some powerful AI technology. But continuous chemical manufacturing processes like ours produce uniquely complex and messy data that their algorithms simply didn’t understand. So, the end results either made no sense or were clearly inaccurate.”

Finding The Root Causes For Actionable Insights That Make Sense

To solve complex continuous manufacturing process inefficiencies, one must first pinpoint the specific problem.

To begin, an automated root-cause analysis can reveal the hidden causes of a manufacturer’s losses. Based on these results, the manufacturer can be provided with predictive recommendations to reach their optimal process settings to avoid losses and maximize capacity. Finally, proactive alerts will enable their manufacturing teams to prevent losses before they occur.

In the end, the company’s process experts were finally able to pinpoint the primary causes of their yield and quality issues, and were armed with the capabilities to maintain optimum process settings and prevent similar problems in the future.

The end results were indeed impressive: a 65% reduction in the amounts of toxic side products – which translated into annual savings of one million euros for that single production line.

Download the full case study to learn how they did it.

The post How a Global Chemical Manufacturer Reduced Quality Losses by 65% appeared first on Augury.