Intelligent asset management and the race to Zero D

In an earlier post, IBM industry expert Scott Campbell talked about how manufacturers are pursuing resiliency and Zero D to stop defects and improve products and service quality. In part two of our discussion, he discusses how mitigating rework can save millions and offers some insights on the value of creating citizen data scientists.

Can you explain the concept of “detect and correct” and the kind of technologies and processes you need to implement to reach that kind of efficiency and reliability?

The idea is if you can detect an issue or defect at the point of installation — using AI computer vision models — then you can correct that defect immediately without it becoming cemented into rework. The example I always use is a dashboard of a vehicle: the average vehicle has over 300 electrical connectors, and many of them reside within the dashboard. These have to be manually connected, because they’re wires and not easily managed by machinery. If a connector is not seated correctly, it’s going to short or it’s going to fail. This means that function won’t work. But if you catch the error at the point of installation — and this is where computer vision models are so important — you can determine if it is it fully connected, partially connected, or if the line technician forgot to connect it altogether.

This detection capability can also be integrated into an overall quality system and/or enterprise asset management system. In the case of Maximo Visual Inspection, it is tightly integrated into Maximo Application Suite for enterprise asset management and performance, while easily integrating into customer quality alert systems. So, when a defect is detected, it can immediately signal an alert on the manufacturing floor to ensure the worker verifies and fixes the issue before it moves to the next assembly process. This immediate alerting is what avoids expensive rework. In the case of connectors within the dashboard, if defects go undetected, the rework fix for a simple connection gets exponentially more expensive, as it often requires dashboard removal and re-installation.

Using computer vision and AI to see the errors before they turn into rework and fix them right then — and in some cases a company is willing to stop the manufacturing line to fix a problem before it gets cemented — is a pretty significant capability. Especially because scrap and defects can cost a company more than 10% of annual revenue.

When people think of AI models, they think of data scientists and the difficulty in hiring expensive resources who understand AI technologies, deep learning neural networks and specialized AI computer vision models. But what IBM has done is made it extremely easy for the subject matter experts (SMEs) — the people that know what they’re looking for defect-wise — to actually create and manage the AI models. We do it through a user interface that requires no code.

It’s literally labeling a few images within a picture as good or bad or any other decision criteria they wish to define. Then the system can provide auto-labeling based on what has been labeled thus far, greatly reducing the workload. Finally, the existing data set can be augmented to create very large data sets out of the original sample size. This provides the data to build models that give predictable outcomes — in most cases, the accuracy is high as 95% to 98%. The result: subject matter experts take control of the actual models without the need for data scientists. This makes adoption a lot faster because companies use the people who are familiar with what the manufacturer is doing on the assembly line. That expertise is also a major contributor to the high-level accuracy of the AI models.

What about the concept of predict and correct? Does that play a role in driving continuous operations?

At IBM, we asked, what if you could increase efficiency, extend asset lifecycles, reduce downtime and costs — all while building resiliency and sustainability into your business?

Predict and correct is fundamental to being able to answer that question.

We’ve made it easier to digitize operating environments by taking the sensor data coming off of assets, and understand at a point in time the condition and operational status of those assets. And it’s a lot of data! A single production line can produce more than 70 terabytes of data each day.

By understanding the asset’s total health in terms of lifecycle and leveraging historical time series data, Maximo can predict when a failure is likely to occur in the future. If you can accurately predict failure well before it happens, you can remediate it. This predict and correct capability plays a major role in delivering and facilitating continuous operations.

You start with Maximo Monitor — capturing data and gaining visibility into what your assets are actually doing. Then you add Maximo Health, which tells you from a lifecycle perspective what maintenance structure you should be looking at and allows a single view of assets across the enterprise. Finally, with Maximo Predict, you can see into the future to be much more prescriptive with your asset performance management. It’s an evolution, but Predict is where the AI models come together to allow a customer to see where there is probability for failure for all of their assets and take corrective action.

We’ve been talking about the auto industry but I’m assuming that any industry can benefit from this.

Absolutely. And it bridges beyond manufacturing. We’re talking about the pursuit of Zero D and resiliency for manufacturing because it aligns so well to Industry 4.0, but the same technology can be used, for example, in travel and transportation. Consider railways and the ability to understand the assets — which are both the railway tracks and the train itself — and looking for potential failure. Sensor data is part of it, but then AI visual inspection can also be used to visually inspect railcars, wear on couplings, wheels, and wiring as just a few examples.

Traditionally, with cargo trains, there are maintenance yards, and the train will pull in and then maintenance people manually inspect the train. They visually ensure everything is okay before they let it go back on the track. But that industry is quickly evolving to provide inspection while the train is in transit. Cameras over the tracks take pictures of the train and provide the results immediately through AI computer vision models. If there are urgent safety concerns for example, the railway operator could stop a train. If not, it could continue on, but the technicians might say, Okay, the next maintenance window we’re going to need to make these repairs. Not only is the inspection much more complete — and can happen with higher frequency — but it is also much more accurate in prediction, because it’s using sensor data as well as visual data to manage the assets. We’re also seeing this in civil infrastructure and with bridges and roadways. There’s just a lot of places where visual data and sensor data come together.

What are some of the issues and misconceptions that an organization might have when it comes to using AI to predict asset health and build a more resilient organization?

From a challenges perspective, the first one is a company that doesn’t use IBM Maximo EAM (Enterprise Asset Management) as its work order system. Often companies believe they can’t take advantage of the rest of our application suite if they don’t use Maximo EAM across their entire organizations. But IBM’s monitor, health and predict solutions can connect to other EAM systems so that companies can take advantage of their operational data. We can also connect to other systems that are gathering the sensor data and we can feed this data into Maximo Monitor. This is important because two-thirds of operational data goes unused. It also eliminates a hurdle a company might have to jump with another provider, simply because their work order system is in another vendor’s application. We can manage within that and still drive value with predictive capability by introducing monitor, health, and predict.

Another typical issue is that each asset has siloed data into its own repository. Getting data across all the assets collected into a single pool can be very difficult and time-consuming. But we can bring connectors via APIs or solutions like IBM App Connect and help customers consolidate data into a single repository. This repository can capture time series data, and then you have a starting point for building resiliency and sustainability into your business by extending asset lifecycles and reducing downtime and costs. Once you’re positioned for intelligent asset management — and building resiliency and sustainability into your business — you can reduce operational costs up to 25% and increase uptime and availability by 20%. Those are results that no one objects to.

Source: ibm.com

Continue reading

Industry 4.0 and the pursuit of resiliency

Downtime can cost a manufacturer upwards of USD 21,000 per minute. Fortunately, AI has evolved to accurately identify issues and take action. This advanced technology allows companies to easily add intelligent “eyes” to their operations with standard mobile devices — the same smartphones and tablets that you’re using right now. All to quickly identify defects in production outputs as well as remotely monitor assets for potential disruptions.

I talked with IBM expert Scott Campbell about this AI evolution and his current focus: helping clients intelligently manage their assets with Zero D, which stands for zero defects and zero downtime. Scott has had numerous product management roles within IBM. And almost all of them centered around some type of AI technology. First in financial environments, then with Red Bull racing (where his team used AI simulation to understand race dynamics), and now as the lead product manager for IBM Maximo and IBM TRIRIGA.

What’s the biggest challenge manufacturers face right now?

Every manufacturer knows that there’s a tremendous value if you can eliminate defects and stop rework. If you can keep your manufacturing facility running 24×7 without any downtime, it’s almost a given that there is ROI there. The challenge is how do you actually transition from a reactive environment — which is where most manufacturers are — to a proactive environment. So instead of thinking, we have a problem to fix, how do you think instead, we’re anticipating problems to fix before they actually become problems. The cool thing is, IBM has AI technology that is sophisticated enough to let a company do that effectively. But we have to make sure that it’s trustworthy. When a company looks at all this data, they have to believe in it. Otherwise they’ll go right back to reactive maintenance.

A lot of people talk about Industry 4.0, but I think the big challenge for many manufacturers is how do you even get started? How do you take something that’s transformational and evolve it over time? Because you can’t do this in a big bang approach, or a forklift upgrade approach. You have to evolve it. And you have to start somewhere.

Starting with defect detection is a good way to get introduced into an AI environment that’s fairly easy to understand. It’s pictures, it’s images. You can see the system is doing a better job than an individual can do, and that makes it easier to expand use of that technology. Once you begin to build trust in those results, it’s easier to use machine learning and AI technology for maintaining the assets running on the manufacturing floor. Then you understand the health of an asset, you know hey the odds are really high — a probability of 85 to at 95% — that this asset is going to fail sometime in the next 45 days, so let’s do something about it. 

And manufacturers are moving toward this?

Oh, yeah, you’re seeing it across the board. There’s a big North American auto manufacturer using AI visual detection and predictive monitoring, and they saw immediate results. It’s incredible how quickly they were successful just running a simple pilot. They found 30 defects in the first 30 days, which isn’t that big of a deal. But they were looking at one single connector in one point of their installation, tied to one specific problem for them. When they expanded that to multiple locations on their assembly line, they found up to 200 defects a day. So in the very first month they gained a USD 1.8 million savings on that one manufacturing line.

There are two parts to the Zero D story. Visual inspection and asset performance management (APM). Visual inspection uses computer vision models focused on quality inspection. APM uses machine learning models based on time series data to determine health of assets and probable failures in the future. Toyota is using Maximo Visual Inspection, and now they are also using the Maximo Asset Performance Management (APM) suite. They tested Maximo APM on some of their machinery that does liquid cooling and found that was another problem area for them. By implementing the software into this pilot, they are now able to monitor the asset health 24×7 and predict probability of failure in the future. It is the foundation for them to shift from being reactive and cycle-based, to practicing a proactive, reliability-centered maintenance strategy. This will be transformational for their entire organization.

Those are just two examples of where Industry 4.0 and how intelligent asset management has started to gain traction. Of course, there are lots of others, but those two examples are true showcases for transformational manufacturing processes.

Does adopting Industry 4.0 bear out all the way down the line to the customer?

Yes, it does, especially on two fronts: quality and meeting demand. For Toyota, quality is mission one. Fewer recalls and less warranty work (compared to other vehicle brands) drives customer loyalty, not to mention reduced costs for rework.

Then when it comes to meeting demand, it’s estimated that downtime costs on average about USD 21,000 per minute. That means within an hour, you have a million-dollar problem. And if you can’t meet the demand, somebody else will. There’s loyalty in car buying, but there is also availability, especially with the chip shortage.

When it comes to defects and downtime each topic seems big enough on its own. Why not tackle them separately? Why do you advocate handling them both at once?

Either one of them is critical. But you achieve true transformation when you attack them both at the same time. Because no matter how high your quality is, you can’t meet demand if you have downtime. Conversely, even if you’re super effective in your manufacturing processes but your quality inspection is poor, you’re just adding to your scrap heap or your rework at a tremendous pace.

That’s why it’s the combination of AI-based visual inspection for quality and asset performance management for predictive repair that lets you increase quality and production efficiency at the same time — and that helps build a sustainable and resilient business.

Do you have any other hard numbers around the savings that an intelligent asset management program could bring to a manufacturer?

Of course, this approach is applicable beyond the auto industry. It’s just a very good use case that folks understand. If you look at the rework of a defect — and it’s important to distinguish between a defect and rework — a defect can occur in the production line, but it only becomes rework if it goes undetected through final production. If you catch it, and fix it, before it gets to the next stage of the line, it’s no longer a defect. We emphasize detecting and correcting at the point of installation.

If a defect turns into rework work — which means if it’s either caught in final inspection or somewhere down the line, even potentially by the customer — then it’s about USD 300 per incident. So, if you think of that North American auto manufacturer who found 200 defects a day, they saved USD 300 multiplied by 200 defects multiplied by 365 days. That’s how you hit very large numbers very quickly in terms of saving.

Can you also talk about the savings from not over-maintaining an asset and only performing maintenance when it’s actually needed?

When our customers understand “$21,000 a minute,” they tend to create very rigid maintenance schedules. The problem is they have no knowledge of what actually needs to be fixed. It becomes hey we’re going to check everything once a week. There’s the idea that frequent maintenance schedules are cheaper even though it’s overkill.

But with an APM platform, you can reduce your maintenance and improve uptime — all at the same time! It is prescriptive in terms of understanding where you need to actually apply resources. It provides 24×7 monitoring of the health of assets, can detect anomalies before they become critical issues, and can predict the probability of failure in the future. Technicians are no longer tied to calendar-based scheduling. Because now a company has the data that indicates these assets are just fine and they’re not going to fail for another month, several months, or even years. This means technicians, who are becoming a scarce resource, can better schedule their time. And companies can utilize technicians much more effectively in the areas that have the highest value, based on data they can trust.

Source: ibm.com

Continue reading

Boosting engineering efficiencies with digital threads

As the demand for agile development grows, so does the demand for digital threads. But what are digital threads, and what is their value?

Digital threads enable a product’s development through its lifecycle domains to be digitally traced. The tracing can be in either direction, upstream or downstream in the product development lifecycle.

“Digital” means that all the paths are electronic — no one is searching through filing cabinets looking for requirements or test documents. “Thread” means there is a traceable path that anyone can follow between processes and across data domains.

The value in digital threads is the ability to quickly trace an entire development process, digitally identifying causalities between the processes and ensuring that their datasets are consistent. When there is a product quality issue, the digital thread allows for the part or subsystem to be traced back through its lifecycle. Original design concept, requirements, test cases, quality checks, and signoffs are all available to the engineering team diagnosing the quality issue. If a requirement needs to be changed, that change’s impact can be determined digitally by its logical thread.

The need for digital thread enablement has been amplified by growing market pressures for development to be more agile, as well as the increasing software integration in “smart products.”

Digital threads in agile development

The need for agility is especially clear when dealing with supply chain issues. Companies need to know the dependencies in product design for every part and subsystem.

This is important for accelerating reaction time, for instance, when a delay from one vendor forces the substitution of another vendor’s part. Leveraging a digital thread helps companies automate changes and execute them faster. Manually managing changes can make an organization subconsciously adverse to change, which is completely averse to the concept of being agile.

Digital threads in software integration

The increasing integration of software introduces a new market model for many companies, and it increases the need to track and maintain their products in market. Software offers an attractive potential revenue stream through in-market upgrades and enhancements. Maintaining a digital thread helps engineering teams automate necessary processes and leverage data analytics for better insights to proposed changes, such as optimizing the release of fix-packs, retro-functions, and incremental enhancements.

The digital thread is incredibly valuable in the development process. It enables decision-makers to analyze the impact of changes before they are made, establish cross-functional KPI’s for measuring readiness and progress, or more readily respond and report on cross-functional compliance requirements.

IBM designed the Engineering Lifecycle Management portfolio on a digital foundation that not only leverages a single view of digital development data, but also leverages an industry-standard, open exchange format of Open Services for Lifecycle Collaboration (OSLC). The IBM Engineering requirements, test, and workflow management tools, as well as the systems/software modeling tool, all leverage this digital foundation to establish traceability across the development environment. This ELM digital foundation can be leveraged via its OSLC architecture by third-party tools and other domain applications, further extending the capabilities of ELM’s digital thread services. This would allow a manufacturing management system to extend the digital thread traceability across development into the actual product’s manufacturing.

Source: ibm.com

Continue reading

Hybrid cloud for manufacturing equals resiliency, agility and flexibility

The manufacturing industry, spurred by COVID-19, is increasing its automation efforts with cloud technologies that offer resiliency, agility and flexibility. However, on the shop floor, slow progress to modernize infrastructure and applications prevents companies from achieving gains in productivity and operational efficiency. One of the key reasons for this lagging adoption is the reluctance of plant managers to implement digital technologies. This exists due to concerns around security, latency and resiliency, as well as a lack of understanding about new generation cloud technologies and how they can coexist with legacy applications.  

How manufacturing is evolving

Manufacturing 4.0, often interchanged with Industry 4.0 or Smart Manufacturing, is a term referring to the paradigm shift comprised of major technological innovations in manufacturing. These include cloud and artificial intelligence (AI) technologies that address manufacturers’ challenges and build confidence through application and infrastructure modernization efforts. This tech maturity unlocks benefits through intelligent automation using, but not limited to, the following: Sensors and digital transfer of data, advanced robotics, Internet of Things (IoT), mobile services, 3-D printing and data analytics.  

Cloud computing is paramount for manufacturing companies undertaking this shift — especially in the areas of data processing, data storage and enterprise resource planning systems. Hybrid cloud brings cloud directly to the manufacturing facility and provides benefits such as on-demand computing failover and auto-scaling, while ensuring that the operations keep running even if there is a connectivity failure. It also allows control decisions to be made in real time.  

Scalability is the key for success

The manufacturing industry is on the forefront of technology. Consider an average steel plant, oil refinery or an aluminum smelter, all of which had considerable industrial automation built at the equipment level long before we started talking about IoT. For these types of manufacturing, there is a great deal of legacy technology investment in the plant.  

Much of the data generated at the equipment level is used for running the plant and improving operations. Coupled with the fact that many large facilities are in remote areas — often in proximity to mines or oil well. In remote facilities, telecommunication infrastructure and skill availability are more often than not an issue. This inhibits the manufacturer from taking full advantage of Manufacturing 4.0.  

Open integration and supporting microservices enables an architecture that is flexible and scalable. This scalability and flexibility need to be accompanied with resiliency from a failover and model-drift perspective. Performance and security aspects cannot be over-emphasized considering the real-time and mission-critical nature of manufacturing operations.  

Hybrid cloud benefits

There are four primary benefits to hybrid cloud in manufacturing: faster cycle time, improved visibility, reduced cost and better management of plant applications.  

Faster cycle time and improved visibility: In typical analytics projects in plants, if data scientists spend three weeks doing the initial analysis, they spend three months trying to discover data, collect data and provision servers. Machine learning and AI workloads running in isolation may not be able to extract the intelligence of out of the huge data volume. When visibility improves, data scientists can access data on-demand and spend more time on the value-add activities of actual analysis and model building.  

To achieve faster cycle time, we need to use AI and ML services offered by public cloud, while addressing the challenges of security across locations, data latency, operational visibility, compliance and regulations. 

IBM Cloud Satellite can incorporate any public cloud service to plant-level data without moving it to cloud. By deploying IBM Cloud Satellite in plant locations or within an existing data center, access to public cloud service is possible. This reduces the cycle time of projects, improves productivity and allows manufacturers to run more projects.  

Cost reduction and improved plant management

In an audit of where data is generated, stored and used, frequent data movement is visible. This movement is wasteful and expensive. Currently, many plants aggregate data before storing in consideration of cost and network congestion. An on-premise database managed by IBM Cloud Satellite solution saves on data egress charges and reduces the management effort of plant IT teams. Better IoT integration and smarter data placement reduces costs, enables faster projects and cuts the hot-data access time significantly.  

An industrial edge (powered by data fabric and IBM Edge Application Manager) helps us to generate relevant and timely insights at scale. Creating Continuous Integration/Continuous Development (CI/CD) pipelines for plant applications provides more confidence to push changes in the operational technology (OT) landscape that reduces the need for a longer shutdown. The iterative learning related to training, deploying, learning and redeploying is especially important in data science and AI application. Enabling agile iterations allows users to improve AI applications and get more value.  

Hybrid cloud for the manufacturing industry is unique given the mission-critical, real-time applications that require low latency and high security. These factors lead to multiple smaller on-premise clouds instead of one large cloud. Hybrid cloud in manufacturing follows a hub-and-spoke model, where each satellite node has a relative lower footprint. A central on-premise cloud will manage these multiple mini clouds which may lead to some variance in standardization across various clouds requiring multi-cloud management technologies.  

Many plant applications are tied to hardware and are critical for plant control. Migrating these applications to new architecture requires special care. There are multiple legacy applications and the importance of these applications may warrant an incremental modernization instead of modernization of all applications at once. Hybrid cloud needs to support bare metals and virtual machines during the journey to containerization and cloud-native computing.  

IBM and Red Hat architecture ensures a journey to cloud that balances the imperative to modernize with the need to minimize disruptions in operation.

Source: ibm.com

Continue reading