IIoT in the plant: three ways to profit
Business goals and organization must align to succeed.
Buzzwords abound when it comes to technology in manufacturing: digital transformation, Industrie 4.0., and the Industrial Internet of Things (IIoT). Regardless of the terminology, manufacturers are aggressively embracing new ways of using technology to achieve business goals like improved productivity and profitability. A recent study reports a 72% increase in productivity and a 69% increase in profitability from the applications of IIoT processes. At the same time, many manufacturers are missing the opportunity to improve their business outcomes simply because they don’t know where to start.
The vast amounts of data that connected manufacturing equipment already collects represents one such opportunity that can quickly pay dividends for manufacturers. This data holds the key to achieving all manner of business value. Yet, despite unprecedented access to this wealth of information, only 3% of U.S. manufacturers are fully utilizing it and other operational data.
For any IIoT initiative to succeed, companies need to focus on two important things. First, they need to clearly identify the business goals they are trying to reach, and second, they must identify the team and organizational adjustments necessary to reach those objectives.
The most effective IIoT projects start with small goals that are measurable and outcome-focused. One example is the ability to accurately predict potential equipment failures before they cause production slowdowns or stoppages. Success with smaller, focused projects can help justify investment in larger projects and provide the learning required for scaling to broader use cases, such as plant-wide equipment optimization.
In parallel, it is imperative to ensure cross-functional collaboration between IT, trusted partners, enterprise-wide organizations, and the people who work with the equipment every day – such as plant engineers, technicians, and operations managers. Getting everyone on the same page helps ensure that the right data sources are available and paired with the right skill sets to extract maximum value from the available information. Management buy-in is another must, to ensure support across the company.
Three places IIoT can help
Three of the most effective ways to put IIoT to work, and lay the groundwork for broader benefits, are predictive analytics, condition-based maintenance, and real-time equipment optimization. But for any of these to yield results, it is important to consider the project from an asset lifecycle perspective.
A well-planned IIoT system uses sophisticated analytics and machine learning to sift through the mountains of data generated by connected factory machines. Using digital models of the equipment, the system is able to run scenarios that help it identify potential failures and quickly pinpoint the root cause when failures do occur. Additionally, these models can provide insights into a variety of issues that can help optimize an asset’s performance throughout its useful life.
There are three key areas to focus on at this stage of IIoT development:
1. Predictive analytics
Employing predictive analytics to enhance things like quality improvement and demand forecasting has been a staple of manufacturing organizations for years. But it can now play a pivotal role in improving a broader range of business metrics, such as machine uptime and longevity by predicting malfunctions or failures before they occur.
Comparing data flowing from equipment sensors with the failure history of like machines, an IIoT system is capable of identifying patterns or behaviors that signal an impending change in the machine’s condition. Other information sources such as environmental conditions or machine specifications can add further context. This allows plant engineers or other subject matter experts to use the results to guide the creation or modification of rules that enable preemptive action to address an issue before it causes a failure that results in a line slowdown or stoppage, or worse, damages the equipment.
The more data the system collects the more intelligent it becomes—allowing expanded automation to increasingly replace manual corrective actions to resolve issues faster. For example, automatically slowing machine operation to minimize damage until repairs can be made at a time that has the least impact on production. Further, this growing intelligence can produce greater accuracy in identifying root causes of problems, and significantly increase first-time repair success by creating detailed repair plans tailored to the specific situation.
2. Condition-based maintenance
Making broader use of valuable machine data, the IIoT system can analyze current and past datasets to create a condition-based maintenance program that recommends service based on each machine’s actual usage and operational condition. Typical maintenance schedules are based on standardized intervals – such as units made, hours operated, or elapsed time.
But because these intervals don’t reflect the real-world condition of each piece of equipment, they can lead to costly over- or under-servicing. Waiting too long for service can lead to unexpected breakdowns that halt production and require expensive emergency repairs. Servicing too often can waste money, resources, and technician time.
The system avoids these issues by applying data analytics to digital equipment models, real-time machine data, historical records, and other contextual information. The resulting insights allow it to detect changes in the equipment’s state, impending problems, and probable causes, as well as formulate service plans tailored to each machine.
Operations managers and plant engineers can then plan maintenance service based on actual asset conditions at a time that won’t hinder production. This can extend the useful life of heavily used equipment and reduce maintenance and repair costs. Optimal maintenance intervals also contribute to smoother operations and reduce, or even eliminate, unplanned downtime.
3. Real-time equipment optimization
Conventional wisdom holds that certain machines just perform better than others. But with real-time equipment optimization, this doesn’t have to be the case. Many factors contribute to variances in efficiency, output quality, or yield—from environmental conditions to a particular machine’s factory configuration. Sensor data provides clues about each machine’s performance, but sophisticated analytics are needed to analyze it quickly and accurately enough to detect equipment inefficiencies, assess variations in production capacity, and adjust to achieve optimal operational performance in real time.
An IIoT system can map the behavior and operating parameters of a high-performing piece of equipment by applying machine learning and advanced data analytics to its digital model. This baseline is made up of equipment settings, environmental conditions, calibration, service intervals, configurations, and other variables.
By comparing this optimized profile to real-time machine behavior, plant engineers and operators can determine the best calibration settings for the equipment to raise production capacity and efficiency. Operators also can make performance adjustments remotely, or the system can make them automatically. Some examples include temporarily reducing output speed if a machine’s operating temperature exceeds a specified threshold or adjusting to changes in ambient humidity.
Using this sort of optimized profile can raise the performance of an entire population of equipment, whether in a single plant or at multiple global locations. As the system learns from the continuous optimization process its intelligence grows over time, which makes it better able to extract peak performance from all connected factory machinery.
Moving forward with IIoT
After identifying a business case that will deliver measurable results, it is a matter of getting stakeholders organization-wide on board to support the initiative and to work with the relevant personnel from IT, operations, trusted IIoT partners, and other organizational groups to make it happen. The adoption of IIoT is a maturity progression that starts with connected equipment and moves from simply monitoring machine data to using digital models, powerful data analytics, and machine learning to enable use cases such as predictive analytics, condition-based maintenance, and equipment optimization.
Automation, through the use of dynamic, rules-based logic, can orchestrate complex actions to interface with systems—such as service ticketing and parts inventory systems—to streamline and accelerate maintenance and repair processes. Further benefits can be gained by performing much of the logic and automation directly on or near equipment, allowing immediate responses such as shutting down equipment to avoid a safety issue, and conserving network bandwidth and storage.
Manufacturers around the world have proven that IIoT is improving their productivity and profitability. It is a straightforward way of capitalizing on the wealth of data that connected equipment already generates to enhance operations across a manufacturing organization.
Dave McCarthy is senior director of products at Bsquare Corporation.