Conduct asset performance management with a software-based approach

Unite separate parts of the organization, allowing teams to work collectively

By Emrah Ercan April 8, 2020

Manufacturers embrace digital transformation to boost operational efficiency, minimize risk and improve productivity. Many accomplish these goals through better understanding of asset performance. The most common metric for measuring manufacturing productivity is overall equipment effectiveness (OEE), which identifies the percentage of truly productive production time. In other words, productive time is where good parts are made at optimum efficiency, without downtime.

First popularized in the 1960s by Seiichi Nakajima, founder of the total productive maintenance system, OEE is a function of a unit’s availability, performance compared to designed capacity and product quality. OEE is commonly thought of as a manufacturing key performance indicator (KPI) in that it provides a thorough evaluation of asset productivity, whether it be for a manufacturing line or entire plant. Quantification, i.e., data, gives production managers greater visibility into where and how effectiveness is lacking.

Addressing these shortfalls in effectiveness is one of the best ways to improve plant productivity. For OEE, equipment productivity loss is categorized as the “Six Big Losses” as follows:


  • Equipment failure
  • Setups and adjustments.


  • Idling and minor stops
  • Reduced speeds.


  • Process defects
  • Reduced yield.

The Six Big Losses provide detail into the factors that undermine manufacturing productivity and guidance as to what specific areas are best targeted for making improvements. In this sense, OEE offers a framework for uncovering the core issues limiting efficiency.

Since the inception of OEE, digital technologies have enhanced measurement accuracy and efficient application of improvements. And, with the ability to unlock insights from thousands of data points, advanced analytical tools like asset performance management (APM) software give operators the ability to make incremental OEE improvements, and thereby, productivity enhancements. Today, APM software offers solutions to optimize OEE even further.

The next evolution in APM software allows manufacturing facilities to collect and integrate historical data to build a dynamic model — a digital twin — that ingests new data to predict the remaining useful life of critical plant assets. As a virtual representation of a plant’s assets, a digital twin is modeled from past performance data, real-time present data, and “future” data supplied by machine learning algorithms and guidance from engineers.

A digital twin is valuable for its ability to detect bottlenecks, facilitate predictive maintenance programs and identify benchmarking opportunities for informing OEE efforts.

On a business level, APM software also can offer greater understanding of real cost-of-production and return on investment, which will help businesses optimize end-pricing to customers, making them more competitive.

Predict remaining useful life

A digital twin built from APM software allows mimicking of asset lifecycles, simulating their remaining useful life. In manufacturing environments, this approach understanding the current state of critical assets and predicting how they will fare in the future. By simulating forward — using historic data trends and current operational dynamics as a guide — a digital twin delivers a view into the future.

By modeling outcomes that would result from changing key parameters, digital twin technology provides important insights for optimizing OEE and driving continuous improvement.

In addition, virtual representation models can be designed for the component level, the system level, i.e., for an entire production line or the process level (the entire manufacturing process). If a model estimates the period remaining before the asset is likely to break down or reach the end of its useful life, engineers can stress-test input changes or take preemptive action to prevent failure before it happens. As adjustments or repairs are made, this data can be fed back into the digital twin to determine whether the future state of the component, asset, system or process thereby changes.

Detect bottlenecks

Bottlenecks impede material flow — and hence, productivity — in manufacturing facilities and constitute the weak points in any supply chain operation. The throughput at a bottleneck determines the pace on a manufacturing line and is the limiting factor for efficiency in a facility.

The root issues of bottlenecks generally include process and machine limitations, such as outdated, inefficient or faulty equipment requiring continuous repairs and resulting in slower production and downtime. However, bottlenecks are not always apparent. The outcomes of a bottleneck — for example, production shortfalls — might be painfully obvious, but where and why it is occurring might not be.

In these cases, a virtual twin assembled from historical data is a tool for uncovering bottleneck sources. By examining trends across the data record, such as equipment performance or production data corresponding to each manufacturing line, facility managers can discover those factors that impact plant efficiency. For example, chronic underperformers can be identified or historical data can indicate how often a line has run out of a certain raw material or packing supply.

What makes bottlenecks difficult to document is that root causes may not be readily apparent. But with a clearer view of the facility, managers can better discover their underlying cause.

Take the example of a brewery. If historical data reveals that a pasteurizer can produce a greater volume of beer compared to what can be handled by the packaging line (the next step in the production process), the bottleneck’s root cause will be evident. But if that packaging line is being pushed to operate at maximum capacity and absorb everything coming from the pasteurizer to meet a target output, the brewery might be more likely to see the bottleneck as arising from equipment failure on the packaging line. This scenario might also result in shortened asset lifecycles. With access to this insight, a facility manager might realize the best course of action is to invest in enlarging the packaging line rather than running equipment at its design limits.

Predictive maintenance

Creating a virtual asset representation from historical data and simulating performance forward under varying scenarios allows operators to anticipate exposure, wear on components, risk factors and where failures are most likely to occur. Simulations developed from historical knowledge also can include contextual data about an asset’s maintenance schedule — corrective maintenance performed, repairs and replaced parts. This can be merged with further information such regarding performance records, operating environment and newly available IIoT data collected from the physical asset.

By combining all this data, predictive models can advise as to maintenance actions for minimizing unplanned downtime and potentially eliminating the need for fixed maintenance schedules.

Benchmark for OEE

Digital simulations also are used to conduct benchmarking between sites and assets for optimizing OEE efforts. By benchmarking real-time production and performance metrics, regional managers can make accurate comparisons at the asset, system and process levels.

Benchmarking data can be used to compare chemical treatment programs in different regions or the performance of cooling towers, boilers or specific water treatment technologies across multiple sites. In a packaging scenario, benchmarking data provides plant managers with greater visibility into output, cost and investment performance of each packaging line within a facility.

With performance, availability and quality measurements, manufacturers can generate OEE benchmarks of their assets and production lines to understand which areas of their operations are the most productive and where improvements need to be made. These evaluations also can be extended to the entire facility, allowing OEE benchmark comparisons at a larger scale.

Improve financial performance

OEE also offers tremendous value as a business KPI. Applied as a business metric, OEE enables companies to view their operations in financial terms, helping them understand where to deploy resources for improving cost performance.

Here, APM software provides benefit by delivering deeper insights into how improvements can reduce manufacturing costs, increase profit margins and enhance ROI. Armed with more detailed cost awareness in terms of their business operations, companies can better optimize product end-pricing to advantage.

However, recognize that achieving efficiency gain in core operations and improved financial performance does not follow from simply adopting APM software and targeting OEE enhancements alone. Companies must first focus on enacting mature APM software programs based on connecting systems and technology across the enterprise.

As the “glue” to bring previously siloed areas together, such as enterprise resource planning, digital supply networks and enterprise asset management systems, the value of APM software lies in its capacity to unite separate parts of the organization, allowing different teams to work collectively as one unit. Applied in this way, field-level results drive enhanced performance, which translates into financial results through topline growth, cost reduction, safety, quality and capital efficiency.

Author Bio: Emrah Ercan serves as global director of digital solutions at SUEZ Water Technologies and Solutions and is a member of the company’s innovation and digital board of directors. He is responsible for strategic direction, commercialization and development of the company’s digital solutions. Previously, he served as vice president of strategic initiatives at GE Oil & Gas and GE Digital, where he established an internal incubator to fund early-stage digital oilfield ideas.