In volatile markets where growth is coming from favorable global economic momentum, there are very few levers to drive differentiation and profitability. High-volume and low-margin business means that managing margins from within means a focus on efforts to avoid unplanned downtime, optimize maintenance spend and resources, and mitigate production losses.
In today’s current environment, many process manufacturers are struggling to take a step forward from their reactive or “firefighting” maintenance cultures that are driving inefficiencies. To climb out of this rut, process manufacturers must take a strategic approach to asset management. When it comes to driving cost efficiency and reliability improvements, leading adopters are letting risk guide their strategy and decision-making.
The effectiveness of the maintenance and reliability (M&R) function is particularly critical to the managing of margins. In short, maintenance and reliability groups ensure that equipment functions as needed to meet production requirements. This means that M&R is a major factor in production capability. And, without production capacity, the business fails to produce quality product that is delivered to the customer on time.
Predictive maintenance can be a key enabling technology to make your M&R dreams come true. Most industrial organizations today know that they must move ahead quickly to make this tool a reality.
Applications for predictive maintenance are endless. Process manufacturers should look to stable, reliable solutions that offer industrial advanced analytics with a troubleshooting component that enables engineers to rapidly identify performance issues by mining insight from sensor and production data. Here is where digital twin technology can add significant value from an analytics perspective.
There are many definitions of a digital twin, but a simple one is: Digital twin is commonly defined as a software representation of a physical asset, system or process designed to predict, diagnose and forecast asset failures through real-time analytics to deliver business value.
A digital twin helps companies to:
- Understand the past: It tracks historical context and performance data. This is useful for correlating variables or running machine learning algorithms.
- View present conditions: The digital twin is regularly fed with sensor data, through IIoT connectivity, to detect anomalies and improve model accuracy.
- Predict the future: It synthesizes and contextualizes historical and real-time data to give insights into potential future states (e.g., whether an asset is heading toward a failure along with the consequences of that failure and mitigating activities).
Seamless connectivity, rich visualization and predictive maintenance realized through digital twins enable users to analyze operating scenarios, quantifying the impact that operational changes will have on key performance metrics (KPIs) and identifying causes for performance variation.
Additionally, to support the full IIoT value journey, digital twins can provide capabilities from simple calculations to predictive machine-learning models to real-time optimization.
Industry professionals responsible for managing assets are faced with numerous challenges in designing cost-effective maintenance strategies for equipment. Intelligent approaches for identifying optimal processes and systems ideally combine data-driven modeling of known degradation mechanisms with field expertise of asset performance and plant operations.
The ability to construct such models of assets and systems — and using predictive maintenance to affect change — requires understanding of many important features, including:
- Failure modes and risks: Understanding the critical risks associated with a system. Not all failure modes are equal, and it is important to understand the differences and complexities of different patterns.
- Mathematical concepts: Understanding the mathematical and statistical concepts required for data-driven model construction.
- Alignment with strategic goals: Understanding how to align model output with strategic priorities such as maintenance costs, spare holding levels and risk of unplanned system downtime.
A comprehensive predictive maintenance solution provides the analytic building blocks to understand the effectiveness of models, diagnostic rules for multivariate analysis and the ability to quickly train and optimize models with connectivity to real-time data sources. All of this should be enabled by a rich equipment library to quickly deploy and scale flexible out-of-the-box models specific to equipment classes and its operating context.
Plug-and-play connectivity to time series data sources and automation systems make for faster configuration. Built-in support for data quality makes real-time data cleaning and validation easy.
Clean data is a must-have
Bad sensor data that feeds the digital twin or other analytic software can mean downtime, possible compliance issues and sometimes safety risks. In addition, this bad data can affect other continuous improvement programs slowing digital transformation. Industrial organizations need to have good data that can be leveraged for operations.
With the combination of clean data and predictive analytics software to monitor equipment health, process manufacturers can realize many benefits:
- Reduced downtime: Sensors are often used to provide indications that equipment is running correctly. Incorrect readings can lead to equipment failure or damage. Early detection of a sensor that is no longer giving accurate or consistent results can provide advance warning that enables maintenance to replace or recalibrate the sensor before the worst happens.
- Improved product quality and compliance: Sensors are often used for measuring the results of a product or to ensure the ambient surroundings of a process are within specification. If the sensors used to measure the product or environment are not accurate or functioning correctly, it can lead to a product being out of specification. Providing warnings can reduce costs related to product recalls or scrapped product.
- Increased data quality: Ensuring data quality in downstream analytics is part of IoT-fueled improvements. If the intent is more advanced use of analytics for a process, the need for ensuring data quality is critical.
Determining where to deploy the technology
Through the lens of minimizing risk comes a significant opportunity for process manufacturers to converge both preventive and condition-based maintenance activities to drive cost optimization and reliability improvements. This approach enables a quantitative risk assessment methodology to be the driving factor for determining the right recipe of preventative maintenance, predictive models and tactics that should be applied to an asset or system.
Technologies like asset performance management — using predictive maintenance as enabling actions — are equipping industrial companies to sustain these methodologies, not only to develop and deploy comprehensive asset strategies but also to track the effectiveness of preventive maintenance and emerging threats from condition-based anomalies.
The criticality ranking of equipment can be used to select the type of analytics strategy to deploy.
- Critical equipment can be characterized as the heart of the plant and warrants a high level of predictive monitoring with protection and online analysis. Companies should consider optimizing this piece with a financial commitment because these assets affect financial performance of the company.
- Mid-criticality equipment warrants creating trends using the plant historian for performance monitoring or scanning type of device for mechanical monitoring.
- Low criticality equipment is monitored at a set frequency using technologies that measure vibration, oil and motor measurements, etc.
- Non-critical equipment should have the oil changed and bolts tightened at regular intervals but should otherwise operate at a run-to-failure strategy.
Bottom line
One example of how quantitative criticality assessment can inform a strategic approach to maintenance strategy and technology deployment is demonstrated by a company that was looking to reduce its annual maintenance spend while improving productivity. The company decided to focus on its critical subset of assets using predictive maintenance technology to model the expected performance of each individual asset and its failure modes. This analytic measured the functions of the asset against actual conditions of various measurement points leading to early identification of issues. The early warning allowed significant time to plan maintenance activities to avoid functional failures.
The company had two major “catches” in the first six months of its predictive maintenance solution deployment, which helped to avoid approximately 130 hours of equipment downtime. The company estimated that the annual savings it could realize from this solution would be approximately $4.5 million, and the return on investment on the project was only eight months.
Preventive maintenance, data analytics and predictive maintenance strategies can help process manufacturers improve performance and reduce costs. But companies must commit to making this solution an imperative in their digital transformation strategy.
Jared Hartness is a senior industry leader for process industries with GE Digital, working with the company’s Asset Performance Management suite of solutions. Prior to joining GE Digital, Hartness began his career with Meridium, serving in capacities ranging from product marketing, marketing programs and digital marketing for its asset performance management software solutions.
