In a world where more systems are becoming interconnected, we’re seeing the Internet of Things tear down communication barriers in home, the workplace, and at work sites among field service companies.
Because of the business, personal, and industry specific benefits that come with devices capable of talking and interacting with one another (and central operating systems), the growth of IoT has been consistent and is forecasted to rise considerably in coming years.
According to Statista, there were 15.4 billion interconnected devices in 2015. Today, the number of devices have reached just over 23 billion and are charted to reach more than 75 billion by 2025.
McKinsey states that globally, an estimated 127 new devices connect to the Internet every second. In the same report, McKinsey approximates that the IoT could have an annual economic impact of $3.9 trillion to $11.1 trillion by 2025.
Granted, many of today’s connected devices are considered to be consumer technology but with advances in asset management through platforms like SAP, that’s changing sooner rather than later.
For one, Business Insider estimates that the industrial sector, through the onboarding of IoT and connected devices/machines, will invest upwards of $70 billion into IoT solutions by 2020. That’s an increase from $29 billion in 2015.
Cisco recognizes the growth of IoT in the industrial sector as well. While consumer devices represent a large share of interconnected tech, Machine to Machine (M2M) connections will represent nearly half (46%) of all interconnected devices – worldwide – by 2020.
Why the Push for Interconnected Machines and Technology
Right now there are machines operating outside of spec. If it were closely inspected the indicators would likely be visible and obvious to a trained technician. Of those machines you can guarantee that some of them are operating dangerously close to failure.
Something will break, the machine will shut down, and production and workflow will grind to a halt. The entire process bottlenecks until someone can arrive to diagnose and find the fault, find a solution, order the necessary components, and get repairs made.
That happens every day to manufacturers around the world, and that’s why interconnected devices and M2M connectivity is gaining popularity.
Among interconnect machines virtually any system with sensors and the ability to communicate can be programmed as part of a network to monitor itself. Through that monitoring it can automatically respond to changes in its environment whether that’s making adjustments according to programmed responses or signaling for human interaction.
Without interconnected machines and the ability to manage networked assets to scale, the only other options are run-to-fail or cost-prohibitive labor-intensive monitoring.
This is what asset management looked like yesterday (and today) for many manufacturers:
- Untrusted and disparate asset information
- Extremely limited analytics and data capabilities
- Reactive maintenance (fix it when it breaks)
- Shortened life cycles that force selling and purchasing new equipment more often
- Maintenance plans based on budgets and not assets
- Physical media for documentation, instructions, and work orders
That’s the industrial sector we’ve come to know for decades. Is that where you want your business to stay?
Not when it’s clear how digital asset management can so dramatically change operations:
- Access to volumes of operational data that can pulled in real time with simulations
- Collaboration across a network of professionals on site, at the manufacturer, vendor, etc. for a single source of truth
- Longer lasting product life cycles with predictive and prescriptive maintenance
- A maintenance strategy built around cost, risk, and performance – the way it should be
- Interactive work instructions with 3D modeling available across an entire network, capable of being updated in real time
All of that points to a predictive maintenance model; using sensors and connectivity to provide the right data at the right time so decisions can be made to reduce expensive bottlenecks (and system failures/damage).
Reactive to Predictive – Ending the Run-to-Failure Model
The Internet of Things is leading to increased use of on-condition and predictive maintenance strategies.
While they are still relevant in some areas of industrial manufacturing and operation, reactive and preventive maintenance do very little to guard against unplanned equipment downtime and the ensuing high costs of repair.
By switching to interconnect devices in an asset management network you are able to increase your use of more advanced maintenance strategies. That, combined with network-accessible data and modeling, and costs can be reduced considerably.
Case Study – Lockheed Streamlines Operations and Diagnostics with Predictive Maintenance
Aerospace and global security company Lockheed Martin wanted to move away from old methods for maintenance. Technicians manually assessed and tracked damage by applying transparent film and tracing reference points like seams and fasteners, and cross-referencing those with repair history. It was a cumbersome, costly, and time-consuming process that left a lot of room for error.
By switching to an asset network, Lockheed Martin was able to take advantage of big data and 3D modeling to streamline its damage and assessment process across F-35 and F-22 fighter planes.
More specifically:
Increase Operational Availability of Equipment: When an aircraft lands, maintainers on the flight-line can connect to the database and immediately determine whether the aircraft is flight-worthy.
Work More Efficiently with Fewer Personnel: Using a streamlined process, maintainers can reduce the time required to document, assess and repair damage. In the case of the USAF, because the F-35 and F-22 solutions have similar workflow, aircraft maintainers can easily transition between the two platforms.
Capture Data More Accurately: Providing maintainers with the ability to visualize and accurately represent aircraft damage on a 3D model reduces the probability of maintainers making mistakes, which translates into safety for pilots.
Caterpillar is another brand utilizing SAP asset network for predictive maintenance.
While other companies operate their equipment, Caterpillar is collecting data from its connected products and turning that data into actionable insight. Not only do operators know precisely when to tag equipment for maintenance, Caterpillar can also use that data to prevent failure.
This level of predictive maintenance from the manufacturer allows Caterpillar to spot potential issues and send replacement parts, maintenance routines, are alerts to a company well before the equipment malfunctions.
Approaches to Predictive Maintenance
The way Caterpillar monitors its assets on the network is one of many approaches to predictive maintenance, all of which have a singular focus: identify an issue before total failure can occur.
Data is critical in predictive maintenance but it’s not a new thing. Data takes a lot of forms as does predictive maintenance – technically, finding a machine hot to the touch when it’s not supposed to be is a form of predictive maintenance.
You know it’s doing something it’s not supposed to be doing and it if it’s not addressed then there’s a problem.
But the window to address the issue is so small, with a high potential for failure, that you’re practically in reaction mode to avoid catastrophe.
The more data you can access and have available in a timely manner, the more time you have to respond. This data-driven approach to predictive maintenance is what provides the greatest flexibility; you can dynamically plan maintenance events and you gain the control/ability to change unplanned events to planned ones.
The Scope of Predictive Maintenance
The Internet of Things, asset networks like SAP, and predictive maintenance will not work without data. Predictive maintenance requires far more than just having access to vast amounts of data and analytics.
You need to be capable of obtaining value from that data.
It’s not uncommon to encounter maintenance managers who control terabytes of data only to find that most of that data is unusable for predictive maintenance.
For example; when data sources from multiple sites are logged for review but the data is missing timestamps, or when machine condition data is gathered but the state of the machine isn’t recorded or can’t be linked to an event – that data is likely unusable.
Since everything is interlinked (machines, sensors, and products) there’s an abundance of unstructured data providing the status of the machine or equipment. This data gets stored along with process and product data. By intelligently monitoring and analyzing that data any company can (practically in real time) start optimizing their processes and spot trouble well before it starts.
But only when there’s a full scope maintenance and service solution in place with a strategy for taking action.
With a full scope maintenance and service solution like SAP, it’s not just about having data. You can create an end-to-end plan from sensor to outcome that includes the data infrastructure, what is read, how it’s read, and the actions to take.
Predicting equipment degradation and malfunction is key to achieving streamlining maintenance and service procedures. When you understand the data you have available, how to use it, and you develop a strategy based on the data then you can monitor your assets and enhance decision making with far deeper insights.
Like Caterpillar mentioned above, it becomes incredibly simple to evolve and execute a predictive maintenance and service strategy that is entirely data driven and leads to optimal performance.
Data and Predictive Maintenance Permanently Evolves the Product Life Cycle
There are a number of key benefits attracting manufacturers and brands to IoT and predictive maintenance, most notably:
- Improving service profitability
- Reducing maintenance cost
- Increasing asset availability
When you expand your view to include the entire lifecycle of the product you begin to see how big data, interconnected machines and devices, and predictive maintenance benefit every individual touch point along the lifecycle.
In the following image from SAP we can see how value manifests from start to finish when the right data is leveraged:
- R&D can improve the reliability of future product iterations
- Manufacturing can be improved to produce higher quality components
- Service level agreements are better managed with improved response times, less wrench time, and faster recovery
- Maintenance costs are reduced
- Customer satisfaction is improved
- Asset up-time is increased
When data is properly prepared, filtered, and analyzed it’s an invaluable tool for any enterprise.
Eliminate the Leading Cause of Downtime (and Potential for Catastrophic Failure)
According to a 2017 Plant Engineering maintenance study the leading cause of unscheduled downtime is aging equipment followed up by operator error. And while nearly 80% of manufacturers have a preventative maintenance plan in place, it’s far from predictive.
In fact, more than half still use some combination of spreadsheet scheduled and paper maintenance records/reports.
Without an asset network and appropriate data it’s not just difficult (near impossible) to effectively monitor and predict problems. It’s extremely difficult to track all the failures and easily review historical data to determine when future maintenance should occur.
One run-to-failure example paints a clear picture of how this approach to maintenance can and will get away from operators – because the rate of failure is never zero.
In Disaster By Design, Dave Lochbaum talks about the aftermath of missed replacement/maintenance windows and a reactive (rather than a predictive) maintenance plan pushed the limits on safety in one nuclear plant.
“The replacement condenser tubes had a 15-year service life. The owner originally planned to replace the replacement condenser tubes during a refueling outage in 2012, but deferred that task until a refueling outage in fall 2014. (Yes, 2012 is already two years past the 15-year service life of the replacement tubes and 2014 pushed the tubes even farther past their expected lifetimes.) Worn-out condenser tubes began breaking left and right, and center, and top, and bottom. The operators had to reduce the reactor power level to 50% several times each week to allow workers to find and plug the broken tubes.”
Conclusion
Parts, products, and process are going to break down. It’s happening right now, and it will happen in the life cycle of every product and machine. An asset network affords companies access to all the data they need, at the right time, to dramatically prolong a product while improving processes, satisfaction, and support at every stage of the product life cycle. The IoT is changing the way we do maintenance, but it’s also providing those companies actively using asset networks like SAP with the opportunity to improve at every level.
The secret to sustainability and scalability in the age of IoT is about the waste that never occurs; the replacement parts not purchased, the emergency pages that never happen, the notifications of delayed production and missed delivery that never have to take place, the accounts that are never lost.
Predictive maintenance with an asset intelligence network is a permanent shift to saving instead of spending.