Ever wonder why your Christmas lights seem to have a mind of their own? Then take this jolly – and very educational – journey into the colorful world of making sure your holiday illumination stays illuminated. Augury Solutions Architect Cliff West lights the way!

Twinkling Troublemakers

Picture this: You pull out your carefully stored Christmas lights, excited to deck the halls, only to find a tangled mess of un-festive cheerlessness. Don’t worry – it’s not just you! These twinkling troublemakers have a secret science that makes them more complicated than your uncle’s politically charged Xmas dinner conversation.

Over at Augury’s bustling online community, The Endpoint, I’ve written an article that the season was screaming for: ‘Untangling the Mystery of Christmas Lights: Why They Fail and How to Fix Them’.

The shocking truth is that Christmas lights are like a delicate electrical chain gang. Each light is connected in a series, which means they’re basically holding hands in a long, fragile line. When one light goes out, it creates a domino effect of darkness that can drive even the jolliest elf mad! 

Seasonal Cheer On Hold…

Here’s the nerdy (but cool) part: Each bulb has a sneaky little backup system called a shunt. When a bulb’s filament dies, the shunt is supposed to create a secret electrical bypass, keeping the rest of the lights shining bright. But here’s the catch – every time a bulb is bypassed with the shunt, it increases the current in the circuit, making more bulbs likely to burn out.

The mathematics of light strand reliability is hilariously depressing. Assuming each bulb has a 99% chance of working, a strand of 50 bulbs only has a 60% chance of fully functioning. Stretch that to 150 bulbs, and you’re looking at a mere 22% chance of a complete, working light strand!

It’s A Christmas Miracle!

Pro tip: Invest in a Light Keeper Pro. It’s like a Christmas light doctor who can diagnose and sometimes resurrect your dying decorations. It can detect voltage and even send a high-voltage pulse to revive stubborn shunts.

The moral of the story? Replacing bulbs as they fail is key. And maybe, just maybe, cut those Christmas lights some slack. They’re working harder than you think to spread a little holiday cheer!

So, the next time your lights go out, remember: it’s not user error – it’s reliability statistics with a dash of holiday mischief!

Read the full in-depth article: ‘Untangling the Mystery of Christmas Lights: Why They Fail and How to Fix Them’.

The post A Tangled Tale of Christmas Light Reliability appeared first on Augury.

This global manufacturer produces flavors for the food industry and fragrances used in beauty and well-being products. Deploying Augury in one of its U.S. plants, the company found quick time-to-value when alerted on a Rox 1 fan and motor just days after installation.

Industry—Flavor and Fragrance//
Machine—Rox 1 fan & motor//
Fault Type—Rotating mechanical looseness//

From Baseline to Alarm

Four days after installing Augury Machine Health sensors on a Rox 1 fan and motor, the machine came out of baseline, immediately indicating “Alarm” status due to signs of rotating mechanical looseness. According to Augury’s analysis, the asset displayed an increasing trend in elevated acceleration amplitudes at the driven fan.

Figure 1: Platform showing machine coming out of baselining activity and into Alarm (March 2024)

The Recommended Fix

Augury advised the plant’s maintenance team to move quickly and attend to the following recommendations:

• Slow rolling the shaft to feel for resistance or roughness
• Performing life and endplay checks with dial indicators to verify bearing/housing clearances
• Ensuring adequate lubrication

The Taste of Victory

The plant’s team executed a successful repair during scheduled downtime based on Augury’s feedback. Along with new belts, bearings, and laser alignment, they also secured the base of the asset to the concrete foundation below. 

Figure 2: Reduction in overall velocity post-repair, July 12 onward

Figure 3: Reduction in high-frequency vibration post-repair, July 12 onward

Plant floor leadership estimated slightly more than $18K in costs avoided – a huge win in saving product, maintaining throughput, reducing waste, and using energy efficiently.

More importantly, a win like this goes deeper than costs avoided. It can also be counted in less tangible ways, like increasing worker safety and morale when repairs are planned in advance rather than operating in reactive, firefighting mode. 

Ready to learn how Augury’s Machine Health can help you find stability and predictability on your plant floor? Contact our team for a no-pressure conversation on solutions to your toughest production challenges.

The post Tasting Productivity & Smelling Success: Leading Flavor and Fragrances Manufacturer Avoids $18K in Costs with Machine Health appeared first on Augury.

Every manufacturer wants to improve Overall Equipment Effectiveness (OEE) and associated metrics, but achieving world-class OEE is impossible without real visibility into machine health. In this article, we outline how a predictive approach to maintenance can significantly improve a manufacturing operation’s OEE by maximizing machine availability and performance, as well as output quality.

This article was originally published on October 27, 2020 and updated on November 18th, 2024.

As emerging technologies transform manufacturing, traditional approaches to machine maintenance are less effective. Maintenance and reliability teams that utilize a reactive approach to addressing equipment issues and breakdowns lack real insights into Machine Health. As a result, they can’t make meaningful improvements to OEE.

In the increasingly competitive world of manufacturing, this is a major problem.

What Is OEE And How Is It Calculated?

As the premier metric for measuring a manufacturing operation’s productivity, OEE considers equipment availability, performance, and quality of output to assign a numerical value to the overall effectiveness of a machine or production line. Essentially, that numerical value represents a machine’s actual productivity in relation to its full potential over the course of planned operations. It’s expressed as a percentage.

What Is A Good OEE Score?

A low OEE value means that a production line or individual piece of equipment is severely underperforming, whereas an OEE of, say, 100% indicates that an operation is producing only the highest-quality parts, as fast as possible, for the entire duration of planned production time.

Every manufacturer wants to improve OEE and its associated metrics, but achieving world-class OEE is virtually impossible without real visibility into machine health.

Time For A Change

The typical manufacturer allocates approximately 40% of total operating expenditure to maintenance — a substantial amount for any company. Yet when maintenance and reliability teams operate within a traditional reactive or preventive framework for identifying and prioritizing equipment repair, that investment is often insufficient.

Factory walk-throughs, random equipment checks, and annual shutdowns don’t just consume valuable time; they’re also prone to human error and leave opportunities for machine malfunctions to cause unplanned downtime. Furthermore, the data generated by these methods generally isn’t useful for manufacturers trying to forecast future risk and production output.

As a result, manufacturers employing conventional maintenance practices experience more unplanned downtime, lower-quality output, and more maintenance-related costs throughout the year. Eventually, these negative outcomes add up, putting a manufacturing operation at a severe disadvantage in relation to more forward-thinking competitors.

So what’s the solution?

“Not only does condition-based maintenance allow manufacturers to reduce machine downtime and avoid temporary unplanned outages (as well as potentially catastrophic equipment failure), but it also allows them to accurately forecast future risk of failure based on a machine’s current condition.”

How to Improve OEE

Given the increasingly complex nature of modern factory equipment, manufacturers can’t afford to not continuously monitor Machine Health. Real-time insight into equipment performance generated by sensors mounted onto or embedded within individual machines allows teams to perform condition-based and even predictive maintenance. The advantages of this approach over outdated maintenance and monitoring techniques are clear and substantial.

By using vibration analysis and similar data-collection methods in conjunction with advanced analytics, factory operators can anticipate a machine failure before it happens. When a bearing starts to show signs of deterioration, for example, teams utilizing a condition-based maintenance approach can replace the part before it negatively impacts a machine’s effectiveness.

Not only does condition-based maintenance allow manufacturers to reduce machine downtime and avoid temporary unplanned outages (as well as potentially catastrophic equipment failure), but it also allows them to accurately forecast future risk of failure based on a machine’s current condition. By extension, they’re able to more confidently set production targets and plan future maintenance operations around production goals.

To better understand how a predictive approach to maintenance can lead to significant improvements in a manufacturing operation’s OEE, let’s take a deeper dive into the metric’s three primary components.

“In high-level sports, it’s often said that the best ability an athlete can possess is availability. The same is true of machines in the world of manufacturing.”

The Three Components Of OEE

1. Machine Availability

In high-level sports, it’s often said that the best ability an athlete can possess is availability. The same is true of machines in the world of manufacturing. A piece of equipment that’s not functioning at capacity (or at all) when you need it to be operating at 100% is a financial drain on your business. Unscheduled downtime as a result of machine failure is one of the largest obstacles manufacturers face when it comes to improving OEE. A reactive approach to maintenance virtually guarantees that unexpected equipment failure will be a persistent problem.

Manufacturers that rely on a preventive maintenance approach, with scheduled downtime to address equipment issues that may or may not impact OEE, face similar inefficiencies. During a plant shutdown or other scheduled maintenance periods, mechanics and technicians are typically presented with a task checklist and a target completion rate. In order to meet this target within the time allotted for maintenance, they’ll often mark unfinished repairs as complete. This is why a facility with a preventive maintenance checklist that is close to 100% complete can find itself operating at just 50% efficiency.

Moreover, even the best technicians are prone to human error. During scheduled downtime, these experts will often be tasked with inspecting equipment that doesn’t need repair, resulting in further time and cost inefficiencies.

2. Machine Performance

In order to meet the demands of modern consumers, manufacturers must produce as much as possible, as fast as possible. The problem is that an outsize focus on production speed can actually lead to more instances of equipment failure, ultimately hindering productivity rather than improving it. Despite this reality, manufacturers must maintain certain performance standards if they hope to stay in operation.

Idling equipment and minor stoppages due to operational inefficiencies can become major problems if not addressed promptly. However, factories that rely on manual machine inspection processes often don’t know that a machine isn’t performing at full capacity until it fails completely. As a result, they suffer a loss of production that’s exacerbated by the need to divert human and capital resources to diagnostic and maintenance activities.

3. Output Quality

Large manufacturing operations have little room for error when it comes to production quality — they have to get it right the first time. If they don’t, they stand to lose potentially significant amounts of time and money and could suffer severe damage to their reputation as well.

Excess waste — scrap or output that can’t be sent to customers due to structural or cosmetic defects — often occurs when machines aren’t operating at optimal conditions. Whether this results from normal depreciation, overuse, or some other issue, it’s a loss for a manufacturer.

‘But understanding the actual state of your machines with a Machine Health platform is only the first step toward improving OEE.”

Maximizing OEE With Machine Health

Each of these components can be improved manually over time, but if your goal is to achieve world-class OEE, reactive and preventive maintenance approaches won’t get you there — and they certainly won’t get you there ahead of your competitors. Instead, you’ll need a Machine Health monitoring solution that gives you real-time visibility into equipment status, condition, and performance.

But understanding the actual state of your machines with a Machine Health platform is only the first step toward improving OEE. Being able to use those insights to prioritize maintenance schedules and take direct, targeted action to reduce machine downtime, failures, and waste represents the next stage of facility sophistication.

A Holistic Approach to OEE

Within every manufacturing operation, there is a production and a maintenance unit, and it’s logical to assume that the two might work together to pursue common goals. Historically, however, that hasn’t been the case. In fact, their operational mandates have often put these two teams directly at odds with one another.

Maintenance is typically charged with maximizing OEE, while production teams’ objectives tend to shift based on ongoing revisions to quotas or product specifications. Both units are indispensable, but the emphasis on speed and the immense pressure to meet market demands mean production goals are usually prioritized over machine health and maintenance. Inevitably, this leads to losses in productivity and excessive spending on maintenance.

“While industry leaders have understood and implemented TPM’s principles for decades, relatively recent advances in machine learning and data analytics have allowed manufacturers to truly bridge the gap between maintenance and production truly.”

In the latter half of the 20th century, Japanese manufacturing leaders seeking to remedy this situation developed a methodology designed to create synergy between the two teams: Total Productive Maintenance (TPM). The focus of TPM is on reducing operational costs through the continual improvement of equipment effectiveness, and its creators established OEE as a critical metric in manufacturing.

While industry leaders have understood and implemented TPM’s principles for decades, relatively recent advances in machine learning and data analytics have allowed manufacturers to truly bridge the gap between maintenance and production — in a way that the creators of TPM could only imagine.

The Next Evolution of Manufacturing

TPM makes everyone responsible for OEE and calls for a holistic approach to production and maintenance. When manufacturers have access to real-time Machine Health data, such an approach becomes possible.

A Machine Health data platform gives factory operators insight into key equipment metrics (e.g., speed, vibration, heat levels), enabling them to decide whether a machine requires intervention with pinpoint accuracy. These capabilities allow operations to move from preventive maintenance to predictive maintenance, opening up countless opportunities to improve overall manufacturing productivity and efficiency.

With this data, for example, operators don’t just know whether they have the equipment necessary to meet production goals on schedule; they also know whether their machines can support faster or longer production runs. They can then adjust production line speeds or switch production lines altogether to sidestep equipment problems. Without machine health data, however, operators often find themselves gambling on the unknown.

“On a macro level, Machine Health data can be combined with operational data, allowing manufacturing leaders to optimize facility design to maximize the OEE of individual machines.”

When Machine Health metrics decline, a Machine Health platform will alert managers that maintenance is needed before issues arise. The technicians performing repairs then have access to a vast array of data that allow them to work more effectively and efficiently.

On a macro level, Machine Health data can be combined with operational data, allowing manufacturing leaders to optimize facility design in a way to maximize the OEE of individual machines. While it may seem counterintuitive, the reality is that a single optimized machine can actually cause downstream problems and reduce the efficiency of a production line if other equipment hasn’t been similarly optimized.

The Right Data At The Right Time

For example, consider a bottle-filling machine configured to fill 1,000 bottles every minute. A two-minute accumulation time between the filler and downstream labeler means that the labeler will be shut down for two minutes when label rolls need to be replaced. Suppose the filler is “optimized” to fill bottles at a clip that exceeds 1,000 units per minute but the label roll replacement process hasn’t also been improved. In that case, bottles will be filled faster than they can be labeled, resulting in excess waste.

Without a Machine Health platform, operators facing the above scenario might shut down the entire production line for equipment inspection and conclude that the problem is a faulty labeler. However, with the right data, an operator can avoid unplanned downtime and simply reduce the filler’s speed to reset the line and continue production, without having to spot-check each machine individually.

“In the competitive world of 21st-century manufacturing, these operations will quickly become customer favorites, while those that persist with a reactive or preventive approach to maintenance will fall behind and eventually disappear.”

On The Path Of Continuous Improvement

The benefits of a predictive maintenance approach paired with real-time Machine Health monitoring are both instantaneous and cumulative over time. Manufacturers that adopt such an approach will be able to continuously improve OEE, allowing them to set increasingly aggressive production goals and deadlines.

In the competitive world of 21st-century manufacturing, these operations will quickly become customer favorites, while those that persist with a reactive or preventive approach to maintenance will fall behind and eventually disappear. With early adopters of sophisticated machine health monitoring and data analytics solutions already separating themselves from their peers, the window of opportunity for latent adopters to close the gap is shrinking fast.

Want to talk more about pumping up your OEE? Reach out!

The post How To Improve Overall Equipment Effectiveness (OEE) With Machine Health Monitoring appeared first on Augury.

In the world of manufacturing, a well-organized storeroom is the unsung hero of production efficiency. Why? Because a strong program will help you lower inventory and reduce reliance on “hidden” stores and just-in-case parts. It’s also a critical part of any reliability improvement effort, ensuring that maintenance teams get the right parts in the right quantity in the right condition and at the right time. 

How is your storeroom contributing to operational success at your plant? Managed strategically, a spare-parts storeroom will help maintenance teams become more efficient, improving OEE, raise safety standards, and boost morale. 

As an Augury Presales Solutions Architect, I’ve had the privilege of interacting with many manufacturing sectors and stakeholders at every level, and I’ve seen the good and bad – from plants that lack any kind of storeroom strategy to those leveraging machine health data to streamline inventory efforts and parts ordering. 

No matter where you are in your journey, these best practices will set you on the road to success.

1. Build a storeroom team

• The primary goal of storeroom operation is to balance spare part inventory levels to protect against lost production, with the storeroom team providing for the efficient and effective delivery of parts. 
• Access to the store must be limited to authorized persons: The storeroom should have roles that align with the main functions of ordering, receiving, and issuing parts. While various organizations have different structures, these main functions must be covered to enable the smooth functioning of the storeroom. 

2. Optimize the storeroom environment 

• Be sure the storeroom gives you enough room to avoid congestion–the ability to find parts should be easy and include a defined catalog and storage process. 
• The area should be well lit and have designated working areas that enable the storeroom team to effectively execute their work tasks.
• Employ a regular process to support the 5S conditions in the storeroom.

3. Prioritize spare-part organization and upkeep

• Be sure spare parts and tools/equipment are inventoried, controlled, and available, and regularly inspect and replace them as needed; regularly review inventory levels to ensure they are accurate and in compliance–no hidden stashes of parts in remote locations or technician tool boxes/lockers.
• Establish an area that can help spare parts avoid degradation and maintain like-new condition, i.e., bearings stored to prevent damage, belts stored to prevent exposure to UV lights or from physical storage, shafts on motors and gearboxes should be rotated regularly, reduce humidity to prevent rust on metal components.
• Label hazardous cabinets designated for flammable items.

4. Establish inventory controls

• Establish inventory management practices such as cycle counts, obsolete part removal, non-moving-part level reviews. Document receiving practices with inspection to verify parts match specs.
• Put practices in place to maintain the integrity of the parts and bill-of materials (BOM) lists; implement a process for creating, updating, and maintaining BOMs for new equipment, new parts, and obsolete parts.
• Create an automatic reordering processes, after-hours support, and ordering for emergency work.
• Use machine health data to supplement usage and stocking levels.

5. Create a planned-work staging process

• Create a process to kit spare parts for planned work and pre-stage part kits and tools for planned work.
• Create a process to return unused parts to stores within 24 hours of work completion and a process to dispose or restock parts.
• Use machine health data to plan what work to do and when to do it based on machine and parts condition.

By having the right structure and management practices in place, the storeroom can help ensure the reliability performance for the site. Implementing the techniques listed above can help achieve those ends and make the storeroom a trusted component in the journey to production excellence. You can even start to utilize a partnered solution like Augury’s Parts as a Service (PaaS) program with DSV to get the right parts when needed.

If you’re looking for more great resources on machine health, join The Endpoint, Augury’s free online community. You can interact with or learn from other manufacturing pros, comb through knowledge base articles, and take online courses to keep your skills sharp. I hope to see you there!

The post Manufacturing Storeroom Management: 5 Critical Best Practices appeared first on Augury.

It’s not a dark secret. Everyone working with Internet of Things (IoT) technologies knows it’s tricky to get right. Too many things can still go wrong between reading the data and getting it to the cloud where the AI magic can happen. Luckily, according to Augury’s Senior Product Manager for Connectivity Naama Zarfati, we can now work to solve this connectivity problem – together.

I often say, “Let’s all work together following the same objectives.” (Sometimes, I even shout it – but then in a friendly and motivational way.)

Collaboration may sound straightforward, but it takes a lot of effort, communication, and focus to make it happen. And this need for teamwork is particularly true regarding making IoT technologies more seamless and reliable. We must deal with all the pain points together – with our teams, clients, and across the industry – while following a multipronged approach. 

Why IoT Challenges Are Bad For Both Customers And Providers

To align our objectives, let’s look at the problem. It’s straightforward: customers are unsatisfied when they don’t get the service they pay for. And for them to get this service and be satisfied, the system has to work. 

In the case of Augury, to make it flow for our architecture, the system has to:

1) Collect the data from each machine using sensors that need to be at the right place – and stay there. The sensor must handle all the environmental challenges, including extreme heat, extremely low temperatures, aggressive cleaning regimes, etcetera. Day-in-day-out. 

2) Get the data from every machine to the cloud. First, the data must be transmitted internally from the sensors (via Bluetooth) to a gateway or node. This gateway knows how to talk to the cloud (via a Wi-Fi network, cellular, or LAN). And a lot can happen in terms of data hopping between devices and across protocols, which results in the customer getting no service. 

However, once the data is in the cloud, it’s easy. Algorithms can start crunching numbers and human vibration analysts can check anything that gets flagged. Service is now possible and you are getting what you paid for. 

Ready for the next era of efficiency?

• Register for our event

Understanding The Flow Of Data: Low Service Versus No Service

The IoT problem is an umbrella of issues around what we call connectivity. And these issues can be broken down into three root causes: 

·      network issues 

·      hardware/device issues

·      bad installation issues. 

Now, if we understand the whole flow of data, we’ll understand where the challenges are and what customers need to face daily. 

For instance, a fallen sensor might result in some data not making it to the cloud – it’s just not full service for that particular machine. One fallen sensor does not mean the machine is not being monitored, it’s just isn’t being monitored the way we all want it to be monitored. 

But if a node or several nodes are not functioning, it affects more than that single machine. This can result in no service. And like a pyramid: the closer you get to the cloud via a node and router, the more machine data is involved and the more it impacts serviceability if you lose it.

To be clear, this is a problem for every IoT company that connects between different components.

Two Types Of Pain

Besides the ultimate pain of paying for low or no service, there’s another type of pain: high maintainability. In this case, maintenance workers have to deal with IoT issues daily – for instance, resetting nodes or reapplying sensors. Hence, they are forced to do something other than what they are getting paid for: maintaining the machines. 

So obviously, there’s a strong need for a more robust and self-maintainable system. 

Three-Pronged Approach

At Augury, we were quick to face this challenge. Yes, we can dispatch teams to fix issues as they arise, but this is prohibitively expensive in most cases since our clients are spread worldwide and often in out-of-the-way locations.

Instead, we divided the problem into different subproblems and assigned a cross-functional team to tackle each part.

1) Installation. The quality of installation needs to be on the road of constant improvement. The sensors need to stick to where they have been stuck, and be compliant to the environment where they have been installed. The customer’s network must be up to scratch and be able to adequately host our system and have the whole E2E system running smoothly. Happily, this involves easy-to-track KPIs.

2) Alerts. Often, clients don’t know if and why their network is not working correctly. We need to send more accurate and actionable alerts to our customer’s internal teams to deal with these issues.

3) Hardware IoT. This is another road of continual improvement: better and more robust hardware, improved connectivity infrastructure to provide data-driven clarity to clients, more self-healing, more auto-recovery, more durability when facing harsh conditions, more cost effectiveness, etcetera…

Onward to IoT 2.0

In many cases, these three problems can be solved by adapting what we already have. In other cases, it’s better to start from scratch and reinvent the wheel – or perhaps even look at off-the-shelf products that already have a proven track record.

So, by our teams not only aligning with each other but also embracing best practices and products from across the IoT industry, we can arrive at the next generation of IoT – with more stable service we are all proud of achieving. 

We’ve already come a long way by releasing improvements and enhancements that are already paying dividends. It’s a journey, and one that’s not over yet. But already, we are beginning to bring it to life and it’s making a difference for our customers – and this gives us the confidence to move forward.

And I wish I could say there was a single challenge to overcome. The real challenge lies in the mixture. It’s about improving everything in tandem. And to finally get IoT right, we can only do it together. So let’s make 2.0 happen – now.

Ready for the next era of efficiency?

• Register for our event

The post IoT Can Be Reliable And Easy To Maintain – If We Collaborate appeared first on Augury.

Why is measuring magnetic flux essential for interpreting vibration data related to the health of your machines? “It provides the context you and the AI need to make accurate and trustworthy predictions – without any false alerts,” according to Augury solutions architect (and VA) Andrew Pry. So why is measuring magnetic flux only now becoming a standard part of a VA’s toolbox?

Once upon a time, it was an easy call to know if a change in vibration on your rotating asset was something to worry about. You’d be in the factory, so when you registered a speed change with your handheld, you could do your own inspection or go to the operator and just ask if there were any changes. In other words, you could always check whether the machine is actually operating under the same conditions as when you last measured it. 

But there was a downside: you’d have to be very lucky to be on hand at that moment when a speed change signaled an impending problem that could cause unplanned downtime.

This is the power of constant remote sensing: you can pick up these changes in near real-time and then act in time to solve such problems before they become real problems. 

Welcome to the wonderful world of Machine Health.  

How To Stop Flying In The Dark

But with remote sensing, you don’t know what a change in vibration means. In fact, there are many possible reasons. With factories being noisy places, maybe it was just an overall change in external vibration. Or perhaps the endpoint moved. Or maybe management decided to push production as a prelude to a planned downtime. Or maybe it was those pranksters in Operations playing with their dials again… Etcetera.

This is all stuff you need to figure out when working remotely. Vibration analysts need multiple sources of information to properly interpret an asset’s condition in real-time. But this takes time and effort. It’s much easier if you have a single source to tell you all you need to know to make the best possible decision.

This is the magic of measuring the magnetic flux going through the motor concurrently with your vibration readings. If it’s there, you know the motor is running. As the alternating current fluctuates between positive and negative, you can then also calculate the actual speed of the motor… 

And now you are getting somewhere…

A Single Source Of – Constant – Truth

When I started at Augury a few years ago, we were the only ones measuring magnetic flux. People still needed to rely on their handhelds – and hope their manual inspection aligned with an impending problem so it could be caught. With Augury’s solution, the monitoring was remote, automatic, and constant. 

And with a continuous condition-based monitoring system comes many advantages. Instead of measuring, say, once a month, you’re measuring, say, every hour or less. It also opens up the opportunity to have one sensor giving you all the information you need – instead of relying on another outside source of information, may it be an operator or a whole other system. 

“I would love to describe the whole dance between vibration and magnetic flux. It fascinates me how their signals almost line up – even though they are seemingly wholly separate phenomena.”

The Dance Between Vibration And Magnetic Flux

If I had the space, I would love to describe the whole dance between vibration and magnetic flux. It fascinates me how their signals almost line up – even though they are seemingly wholly separate phenomena. And it amazes me how we can leverage this near-alignment to discover many interesting insights. 

For instance, we can measure slip. And as slip increases, we know the machine is working harder, and its power requirements are increasing. It could be a bad sign. But it could also mean a change in load. By knowing the difference, we can really reduce false alarms. 

And there are all sorts of things we have learned over the millions of machines we’ve monitored. Let’s give two examples… 

Case Study #1: Avoiding The Unsweet Spot

On one level, the measuring of magnetic flux was developed as a way to automate continuous monitoring, but it was also developed so we could track a machine’s vibration against its different speeds to determine whether it has a resonance condition – a whole topic in itself. 

Each machine has its own natural frequency, and if you operate that machine at a speed that coincides with that natural frequency, this causes resonance – which creates a whole lot of vibration.

Happily, we can avoid such situations – and thereby likely extend the life of a machine – by paying attention to any machines that have seemingly big and random vibration spikes. If we pull up the magnetic frequency data to see how fast the machine was running at all those different times, we often discover the machine was actually running at the same speed each of those. Now, we’re sure it’s a clearcut case of resonance. 

Our customers can then choose to avoid that speed in the future, or re-engineer that particular machine to change its natural frequency.  

“Pump happy, we’re happy.”

Case Study #2: Re-Establishing The Sweet Spot 

Many of us know pumps are fussy beasts – and almost human-like in how they hate change. Pumps like to operate in a sweet spot called best efficiency point (BEP). That’s the point where, hydraulically, the pump is in its most stable condition. 

I remember one time, a chiller water pump had undergone maintenance, and when it was turned on, it produced a lot more vibration than usual. Here, you worry about cavitation, a potentially catastrophic implosion on a piece of metal inside the machine. The maintenance team was warned, and they investigated. It turned out the tank levels were not right so they made the required adjustments to put the pump back into its ideal flow conditions.

Thanks to measuring magnetic flux, we could remotely detect an operational change – not just a speed change. Pump happy, we’re happy. 

“What makes Augury technology so powerful is how we’ve built out hundreds of such ‘feature sets’ that help account for our nearly 100% accurate diagnostics. And many of these are thanks to measuring the magnetic flux.”

Now, To Bring The AI Up To Speed

In the name of automation – and making everyone’s job easier – we take these clearly interpreted scenarios we discover and train our AI to recognize them. In other words, we flag those things that the AI should pay attention to, so then, if needed, the algorithm can flag those situations that the maintenance team should look at. 

What makes Augury technology so powerful is how we’ve built out hundreds of such “feature sets” that help account for our nearly 100% accurate diagnostics. And many of these are thanks to measuring the magnetic flux. 

So, consider this a public service message: if you are not yet doing it, measuring magnetic flux is a crucial part of understanding vibration analysis. It’s a particularly powerful tool if you take this measurement at the same time you’re measuring the vibration because you can more easily correlate these two very interdependent variables and gain all sorts of real-time insights.

Fortunately, for both manufacturers and their machines, this approach is becoming an industry standard. But yes, it’s a whole other job to gather enough information to build out an AI that can turn all this information into accurate and reliable actionable insights.

All hail, magnetic flux. 

To learn more about vibration analysis, check out our webinar series.
To learn more about our approach, just reach out!

The post It’s Not All About Vibration. All Hail Magnetic Flux! appeared first on Augury.

Why is Augury’s expanded Guaranteed Diagnostics such a game changer? In terms of industrial warranties, it’s a Mars colony compared to the Earth-bound mundanity of satisfaction guarantees, according to Augury’s Head of Product and Portfolio Marketing James Newman. 

A colleague recently asked what sparked my passion for guaranteeing results. I was surprised since I had never considered warranties to be a passion. Golf: yes. Irish dance: yes. Warranties: not so much. 

But then I realized I had been describing to him Augury’s expanded Guaranteed Diagnostics in terms of space travel and its “revolutionary!”. So, I couldn’t argue. And it’s true, my confidence is boosted when I know something works. And I get mad when competitors peddle false promises in the form of supposed “money-back guarantees.”

It’s also true I am very excited about Augury’s expanded Guaranteed Diagnostics supported by HSB, a Munich Re company. While other warranties are still metaphorically  trying to escape the Earth’s gravity, we’ve already shot past the moon and are happily settled on Mars – with our sights already set on the next frontier.

Sorry, there I go again.

Nuts And Bolts of New Extended Warranty

So, what’s the deal exactly? 

Let’s be clear, this is not a full uptime – or production loss – guarantee. No one is there yet because this would involve taking responsibility for peoples’ actions outside of your control. That’s more the Galaxy far far away movie script.  But neither is it a simple and terrestrial satisfaction-or-your-money-back guarantee. 

Now, we are so confident in our AI-driven capabilities, that we are now also willing to offset part of the resulting production expenses if we don’t flag a machine that ends up failing. 

Augury is offering a singular warranty which includes compensation for unwarned repair events to reimburse for machine downtime and production expenses for Critical Rotating (CR) assets – plus we’re expanding eligibility to include all assets we classify as CR assets. Now, it’s not just about covering any repair and replacement that results from a machine breaking down that we should have been able to find – that’s the moon base we were on before. Now, we are so confident in our AI-driven capabilities, that we are now also willing to offset part of the resulting production expenses if we don’t flag a machine that ends up failing. 

Usual View: Innovation and Mitigating Risk Don’t Mix

But seriously, it’s an exciting proposition. 

For years, technologies – like Machine Health – have promised to reduce your risk and (by extension) change your insurance premium. But in reality, the people who determine your risk – internally or via  your insurance company – are essentially anti-unproven-technology. They really have to see it, to believe it since their job, by definition, is to eliminate risk. And hence, they are the people with the least faith in technology.

And new technologies, particularly when they first appear, can be fraught with unknown risk. So, to be fair, it’s realistic for your risk management team to be anti-technology – or at least until the technology is mature enough that it can actually prove it really does mitigate risk. Meanwhile, the job of insurance companies is not to give away money. And we know many insurance companies take this view very, very seriously – to the point where they stop moving with the times. 

Your production improvement can’t lie in suspended animation
while you wait to arrive at your new planet.

And that all takes time, time you can’t really afford to waste. Your production improvement can’t lie in suspended animation while you wait to arrive at your new planet.

Boldly Going…

Happily, some insurance companies are realizing they have to become less adversarial and even become partners in championing new technologies. And in my opinion HSB happens to be one of those taking the lead in asking how they can do things differently – and then without needlessly opening their portfolio to risk. Of course, they still have to be selective. For instance, you won’t see any insurance companies partnering with generative AI companies anytime soon. Even for this warranty coverage, we are being very clear about what is being offered; this is a warranty on our diagnostics, not new insurance. Nonetheless, Machine Health has proven itself enough over time for a major company to say: “Yeah, we’re good to partner here.”

But this is not for the faint of heart. Compared to other insurance companies, this is an incredibly brave move. They – forgive the Star Trek reference – are boldly going to where no one has gone before.

In other words, Guaranteed Diagnostics guarantees
you can move at warp speed to improve your Machine Health.

Let’s be clear, however, Guaranteed Diagnostics is NOT insurance. That would be easy, like driving your starship on autopilot. Guaranteed Diagnostics is a warranty that Machine Health works. Yes, that’s obvious, but it’s an important point. This is about knowing without a doubt that the solution provider you are partnering with has your back and a stake in the program’s success. In other words, Guaranteed Diagnostics guarantees you can move at warp speed to improve your Machine Health.

Onward and Upward

So, here’s where we’re at – and we have our sights set even further. As mentioned above, we’re expanding our  warranty coverage to more machines. Today, this means all of  our current CR machines are eligible. But as we work  on new types of machines that will eventually become CR machines, this list will only grow..

We are also planning to expand the category from rotating assets around vibration analysis to cover new types of assets as this program continues to grow – going from single colonies to multiple colonies. 

And of course, we are looking very much beyond Machine Health as we start to loop in new data sources to see what types of new insights we can bring together to create more value for our customers… 

And get this: we are offering this coverage at no additional cost for a full year to all new logos we sign on. 

It’s as if we’re paying their transport costs to Mars. How’s that for a confidence booster?

The post To The Stars – With Expanded Guaranteed Diagnostics for CR Assets appeared first on Augury.

Flipping your factory into a modern manufacturing powerhouse requires a strategy that extends into every part of your facility – including how you manage your spare parts inventory.

If you think predictive, prescriptive machine health is just about fixing equipment before it fails, it’s time to think bigger. The actionable insights generated from condition monitoring can be used to optimize asset care—especially when it comes to managing spare part inventory.

Meet the Experts

Adi Segal heads up the maintenance and service department at Bazan Group, an oil refining and petrochemicals company located in Haifa Bay, Israel. 

Sigal Mannheim Katzovich serves as the managing director of DSV in Israel. DSV A/S is the third largest international shipping company, offering global logistics and transport services by road, air, sea, and train.

Working together with Augury, Adi and Sigal discovered a spare-parts-as-a-service strategy, which combines machine health with logistics to better manage production risk. Their just-in-time approach to inventory is an innovative way to save money on maintenance and procurement costs.

#1:  Ditch the Low-Risk, High-Risk Way of Working

Spare parts are a major line item on any manufacturing budget, making it an area ripe for review and innovation. Typically, manufacturers either follow a low-risk or high-risk strategy in the way they manage their spare part inventory.

Low risk = high inventory levels

High risk = low inventory levels

Reflecting on the low-risk strategy, Adi said, “When you don’t know what’s in front of you, usually you purchase more and your shelves are full of spare parts. We found out that they lay on the shelves for more than two to three years, without any use, which is a huge waste of time and money. And from time to time, we can’t even use those spare parts because there’s corrosion or something else that happened to them just sitting on the shelf.”

Conversely, the high-risk strategy means not holding much inventory and hoping that when a part is needed, it will be available, affordable, and delivered on time. According to Sigal, it’s a risk that doesn’t usually pay off. She shared a story where one small missing part cost a customer more than $200,000 due to rushed shipping and the part’s location a hemisphere away.

“Of course, the disruption itself cost millions of dollars,” she explained.  “So the customer is not happy and we work under pressure. This is not a good environment for any supplier.”

#2: Think Differently About Machine Health Data

Bazan’s maintenance team uses machine health insights to get ahead of equipment failure and optimize its assets. However, Adi realized that there was additional value within the data that the procurement department could be leveraging as well. The insights provided a window into better forecasting the need, budget, and timing for spare parts. 

That created an “a-ha moment” for DSV as well. “We realized that we could use the machine health data and insights we got from Bazan to make our whole supply chain better,” Sigal said. “We could buy better, we could deliver better. We have time to talk with other suppliers, with the ocean carriers and the air carriers, to bring parts just in time to the customer to make him happy.”

#3: Create a New “Just-In-Time” Procurement Process

Cracking the code on machine health and logistics meant Bazan’s maintenance and procurement departments pivoted to a new, smarter strategy: just-in-time inventory.

This strategy broke the cycle of reactive inspections, unplanned downtime, and emergency repairs.

“Once AI detects the problem in the machine, there is an alert sent to procurement. We then get the order and we have the lead time to get the best rate and bring it to Bazan at the time it’s needed—not before and not too late—we call it ‘just in time parts’,” said Sigal.

New Process = Big Benefits 

While just-in-time parts help cut maintenance and procurement costs, there are other benefits to pairing AI-powered machine health with logistics. For example, maintenance budgets are typically direct expenses. But predictive, prescriptive machine health extends the life of assets, improves their efficiency, and results in fewer unplanned shutdowns—which add up to significant cost savings for the business.

“I get to control my budget,” Adi said. “I can present to top management where to invest or upgrade a machine because we have spare money now.”

Sigal agreed. “Technology is here to support people and the business. It makes you work smarter, not harder. It’s a benefit–people can analyze data, talk about strategy, and talk about better budgeting instead of running after machines and fixing them all the time.”

The post Strategic Planning for Spare Parts: 3 Essential Steps For Maintenance & Procurement Pros appeared first on Augury.

The three P’s of maintenance – Preventative Maintenance, Predictive Maintenance and Prescriptive Maintenance – are often used interchangeably. But they each have their own distinct meaning. What’s going on? In this article James Newman, Augury’s Head of Product and Portfolio Marketing, decides to set the record straight. “There’s no reason why they can’t get along. That said, Predictive Maintenance will likely go the way of the dodo.”

3 P’s… And 3 Very Different Meanings

Humans are very good at spin. It’s like that quotation that Mark Twain famously used: “Lies, damned lies, and statistics“. And words are like statistics: people will twist them in a way that creates value for themselves. 

Today, you see it a lot with the term “AI” – it’s nearly impossible now to know what’s real and what isn’t. And we certainly see it in how people interchangeably use the three terms Preventative Maintenance, Predictive Maintenance and Prescriptive Maintenance. Let’s put a stop to this.

Instead, we should use these words as they were meant to be used: to build up a more holistic –  and inexpensive – maintenance plan that can benefit both our business and environmental goals. So, to do my part, I decided to break it down one final time into basic definitions so we can once again have normal conversations using these words – and actually make better decisions for the sake of our machines and the people who run them. 

Prescriptive Maintenance: Definition, Advantages, Why It’s Here To Stay 

In short, Prescriptive Maintenance took things to the next level. Using more sophisticated AI algorithms on the data collected from IoT sensors, Prescriptive Maintenance tells you what the problem is in real-time, what to do about it, and in what timeframe. It’s the full diagnostic package. Super handy.

Taking Predictive a step further, Prescriptive basically dooms Predictive to extinction – since Prescriptive gives manufacturers a competitive advantage. And unfortunately, the manufacturing industry is proving itself slow to catch up to this fact.

“Prescriptive Maintenance tells you what the problem is in real-time, what to do about it and in what timeframe. It’s the full diagnostic package.”

But here’s the thing: to run a good maintenance program you need all three P’s – at the moment anyway. At one point, Predictive will get fully replaced by Prescriptive. But until then, it’s likely that you will always have some assets that due to their value, replaceability, or redundancy, you may never be able to financially justify a fully prescriptive solution. And that’s fine. 

Preventative Maintenance: Definition, Shortcomings

Let’s strip it back. Preventative Maintenance means doing something to decrease the risk of a machine failing and causing expensive unplanned downtime. But it is often negatively associated with route-based maintenance – for example, just following the advice of the OEM when the best time is to change your oil on that particular machine.

So it does come at a cost: it doesn’t protect you from unexpected problems and you may be doing more maintenance than you have to. And while organizations have become great at preventative maintenance, there’s a limit on how far it can take you. 

“Preventative Maintenance means doing something to decrease the risk of a machine failing and causing expensive unplanned downtime.”

Don’t get me wrong: you should definitely be proactive and do Preventative Maintenance. After all, you should always change that oil. And it’s definitely a step up from Reactive Maintenance where you are just putting out fires as they flare. 

But you should strive to do it as smartly as possible (more on this later).  

Predictive Maintenance: Definition, Advantages, Why It’s Going The Way of the Dodo

Predicting a problem is awesome. And people have been predicting problems without technology for a long time. For example, that old maintenance guy who can tap the side of a machine with a wrench and go “Uh-oh, we got a problem here”. [Full disclosure: I love those old guys – I am one.]

So Predictive Maintenance is a step up because you know something is up. And in these modern times, instead of some wonderful old guy, you can run some simple algorithms with the data you’ve picked up from sensors – whether they’re handheld or IoT.

“Is it useful if the machine still fails?”

Unfortunately, all this data is telling you is that there’s a bad pattern in the data you collect – that something is different, that something is not right. Naturally, this is super useful. You can send in the techs to dig deeper into the problem and hopefully solve it before there’s a failure. And there’s the rub…

Predictive is not based on real-time but lagging indicators. You  might be too late to do something about it. Is it useful if the machine still fails?

Avoiding The Spin…

You will always have to do some kind of Preventative Maintenance. And if you think you aren’t then you are kidding yourself. But let a Prescriptive program make the Preventative program smarter, letting you know when, for example, you need to change the oil, and when you don’t, instead of just relying on the date or hours run. In this way, you will be able to make it more efficient – so you’re not changing that oil every 500 machine hours, instead of every 200. 

Basically, we’ve gone from the Dark Ages to the Age of Enlightenment in terms of maintenance. We’ve got super sophisticated AI backed by human expertise to create viable Prescriptive Maintenance. In terms of value alone, you’ll want to take this next evolutionary step while still always remembering to listen to the algorithm when it tells you to change the oil (or not to). 

After all, the algorithm has no time for spin.

DuPont + Augury: Driving Innovation with Predictive Maintenance

The post Why The Confusion? Prescriptive Maintenance Vs Predictive Vs Preventative   appeared first on Augury.

For a successful Machine Health solution, you need 5 capabilities – as broken down in the acronym C.A.R.E.S. Each capability in themself is a great step forward. But the real – and measurable – value is in how they interact and build on each other. In other words, you’re missing opportunities if you can’t leverage all 5. For part 1 of this blog series, we put a spotlight on the “C” – Comprehensive Asset Coverage – and how this fits into the larger Machine Health picture.  

Comprehensive Asset Coverage ensures you avoid downtime in the most cost-effective way possible. To make this happen you’ll need the right spread of solutions: from AI-driven prescriptive analytics for your critical assets, down to integrated route-based analysis for less important machines – along everything in between. In short, you need the right level of service at the right time. Meanwhile, with the correct support, you can also go after the ultimate longer-term goals: global scaling and total digital transformation.   

A Warm Blanket Doesn’t Mean Blanket Coverage

“Have no fear. While having Comprehensive Asset Coverage is often described as a warm blanket, it doesn’t mean you need blanket coverage,” says James Newman, Head of Product and Portfolio Marketing at Augury.

C.A.R.E.S.: The Five Required Capabilities of Machine Health

C: Comprehensive Asset Coverage
A: AI: Accurate & Prescriptive
R: Reach At Global Scale|
E: Engagement & Autonomy
S: Services End-To-End

“Yes, people are looking for that security to go home after work and not worry about getting called in to put out another fire. But for that level of reassurance, you don’t need the full package on all your assets – just the right selection. At the end of the day, it’s about not just asset care but also streamlining the processes and costs around asset care,” says James.

“At the core you need the power of AI to solve your major issues, which also enables us to extend that knowledge to assets of lower tiers of criticality, but tht right level of serv ice that they need. The underlying goal is simple: avoid downtime in the most cost-efficient way possible.”

What is Comprehensive Asset Coverage?

To realize the value of machine health, you need a solution that effectively monitors all your required assets and does not add unnecessary overhead.

AI: The Great Enabler of Comprehensive Assets Coverage

“At the same time, it’s not about covering just a handful of assets,” notes Hari Viswanathan, Director of Product Marketing at Augury. “Potentially, you should be able to diagnose hundreds of different machines – especially if you want to scale across a larger manufacturing organization.”

“It’s also not about just sticking a sensor on an asset, collecting data and spotting anomalies. Rather it’s about being able to diagnose the issues. And that’s where Accurate Prescriptive AI shines: it tells you what’s wrong, what to do about it, and in what timeframe.”

“And in terms of efficiency, there are now algorithms that are orders of magnitudes more effective than humans in terms of coverage and flagging faults in critical assets,” says Assaf Barak, VP AI & Physics at Augury. “Algorithms have actually become so strong they can even take on diagnosing new machines without starting from scratch. And over time this will only improve. As we gather more data, we can gain more insights into the assets – to help diagnose the assets ever more accurately. And once you’ve scaled, you can gather further data across many facilities to gain even further insights for your maintenance teams.” 

Root Out Routine: Beware Of Fake AI

“Combining engaged human expertise with AI is where the real magic happens. You can do so much more stuff faster, while missing less,” says James. However, with the rising awareness of the effectiveness of monitoring assets with a machine health program, many monitoring companies are over-claiming their abilities – especially in terms of their AI. 

“This is important when you are implementing new technology: you don’t want to add more processes and steps in their already busy day – you want to reduce them.”

“There are many simple asset monitoring companies that may, or may not, have some form of anomaly detection – which of course is nothing new. And in theory, you can cover your rotating assets in a route-based manner. But no one has the time, people or resources to do this correctly in that manner,” says James. “You simply won’t win. You certainly won’t be able to scale.” 

Empowering Those Actually Covering the Assets

Meanwhile, covering more assets with continuous monitoring, even with great AI, often seems daunting to people who actually have to manage assets. To overcome this fear, solutions must be effective. “It’s about simplifying the professional lives of those on the floor and not making things more complicated,” says Hari.

“You need to prove it works – in terms of avoiding downtime – and that it’s easy to adapt into one’s workflow. This is important when you are implementing new technology: you don’t want to add more processes and steps in their already busy day – you want to reduce them.” 

“In this way, you also need the right support, not just in terms of having easily accessible vibration analysts and customer success managers. It’s a huge change and it needs the right change management to maximize the potential of such a program – from installation to scaling. And this support needs to reach everyone involved, from factory floor to C-Suite. Everyone needs to feel they are backed by an army of experienced teammates.”

“And that’s exactly the warm blanket feeling you need to find in order to properly cover your assets,” adds James. 

“But the real ROI is the journey to digital transformation that full AI-driven asset coverage can set you on. “

Avoiding Downtime: Great.
Full Digital Transformation: The Greater Goal. 

“Naturally, people love avoiding downtime thanks to prescriptive AI,” says Hari. “And they love to track the fast ROI attached to that. But the real ROI is the journey to digital transformation that full AI-driven asset coverage can set you on. You can now loop those short-term savings in terms of avoided asset downtime to actually fuel your digital transformation journey towards long-term savings. In other words, you can take those individual wins to go after the big – and scalable – vision on optimizing production while minimizing waste and emissions. When you kick start that journey up the value chain – you want to be using a solution that grows with you – not one that you need to replace two years into the program.”

“This is why I love my job,” says Assaf. “We are dealing with things we can touch and feel in the real world: machines made of metal, vibrating, and taken care of by professionals with cool hand tools ranging from hammers and screwdrivers right up to advanced laser systems for machine alignment. Then, in a way, these machines end up fusing with sophisticated algorithms that run somewhere in the Cloud. And then, from that huge server, the outputs from these algorithms are aimed back to the customer with their fancy tools to help them fix what needs to be fixed. I can very much connect with this idea: making the circle round.”

The post The 5 Required Capabilities of Machine Health: 1) Comprehensive Asset Coverage appeared first on Augury.