Today, overall equipment effectiveness (OEE) is one of the most common metrics used in manufacturing. The concept was introduced by Seiichi Nakajima in October 1989 as one of the cornerstones in the broader framework for total productive maintenance (TPM) in his book, Introduction to TPM: Total Productive Maintenance.
In the book, Nakajima explains how the primary goal of TPM is to achieve zero defects and breakdowns for each manufacturing machine or asset. Another key objective in the framework is reducing the probability of equipment breakdowns and potentially defective products. Here is where OEE becomes a key metric.
Nakajima observed that reduced breakdowns and defects can help cut costs, improve inventory, and increase production rates. In this context, OEE is an instrumental metric for providing continual feedback on manufacturing performance to the machine level so manufacturers can course-correct their production strategies and gain the greatest productivity possible per machine. The OEE metric is calculated by multiplying machine availability by performance by quality. Thus, by definition, OEE measures the gap between optimal production levels and actual production run rates at the machine level.
OEE is not a panacea. Critics quickly point out that a given manufacturing plant could have a perfect OEE score but suffer quality problems and even miss delivery dates. Moreover, there is a glaring weakness in OEE’s inability to consider manufacturing operations as an integrated ecosystem that can adapt and flex in real time and physics or measurable performance can be learned over time.
While OEE was never designed to provide deeper contextual intelligence, when combined with other metrics and key performance indicators (KPIs) it can play a powerful role in improving shop floor visibility. Additionally, OEE has proven to be a scalable and valuable metric for determining whether a lean manufacturing and maintenance program is going to succeed or not.
OEE’s role in Industry 4.0
For manufacturers, the pragmatic value of OEE, despite its limitations, is in identifying when a given machine has the potential to experience an equipment failure, since companies can prolong the life of production assets by taking prescriptive measures when machinery wear is detected. The majority of manufacturers I’ve spoken with over the last year rely on OEE as one of the metrics and KPIs included in their reporting dashboards for managing production operations.
It’s clear that OEE adoption is growing, especially for tracking Industry 4.0 performance gains to the machine level. In an era of smart, connected machinery, real-time production and process monitoring now goes hand-in-hand with OEE. The more granular real-time data that a machine can provide, the more job- and task-level visibility and control is given to the manufacturer.
Yet, as much as OEE is enabling new levels of efficiency and production capacity across a range of industries, its biggest impact is yet to come as equipment manufacturers evolve their business models.
OEE and the move toward OEM data-based services
As machinery moves toward greater commoditization, OEMs will need to add software services—such as data to support predictive maintenance—in order to grow their businesses and profitability. McKinsey outlines this industry shift in its report, “Digital Machinery: How companies can win the changing manufacturing game.”
According to the McKinsey report, digital machinery means a shift from “hardware only” to “hardware, software, and services” that enables the development of new business models. For example, on top of its own smart factory, an engineering company might also establish a smart-machine-based new business model, enabling new service offers, such as predictive maintenance. Digitization also makes output-based business models possible, which would be the basis for offers such as guaranteed machine performance.
In the emerging world of machinery data-based services, revenues and customer satisfaction will be tied to the data and software services that OEMs provide customers to help them keep their machines up and running and minimize any maintenance-related disruptions to the business. As OEMs provide customers with the services for monitoring and managing their machinery, OEE will play a central role in capturing metrics about machine health along three key dimensions: availability, performance, and quality.
Nakajima designed OEE into TPM to track productivity losses due to downtime (availability), speed losses (performance), and defects (quality). Thus, the OEE equation captures the most quantifiable factors of each area. Further, two components of availability—unplanned downtime (UDT) and planned downtime (PDT)—provide additional measures of how effective a given machine is at any time.
 | Listen to the editors discuss news from PACK EXPO East, including OEE tools. |
In a services-based business model, real-time production and process monitoring provides essential data for getting OEE right to the machine level. At the same time, the use of machinery data for a services-based business model needs to be predicated on an integrated network.
Smart, connected machines can provide the real-time data needed for implementing statistical process control (SPC) and capturing metrics, such as accurate mean time to repair (MTTR), mean time between failure (MTBF), and maintenance, repair and overhaul (MRO) predictions, among others.
Collectively, information is then captured into a manufacturing execution system (MES), enterprise resource planning (ERP), quality management, and other software, for a cohesive picture of every machine’s performance in the context of the customer using it. Then, by delivering this data as a service, machine manufacturers can empower customers to improve equipment availability and performance to maximize their own productivity.
Adding a services-based model may be years away. However, the real-time production and process monitoring capabilities that OEMs build into their machinery now will solidify their ability to leverage OEE in creating and delivering these services in the future.
Conclusion
For machinery manufacturers to successfully scale away from transaction revenue and into subscription-based businesses, they need to fine-tune how they capture, aggregate, and analyze real-time production and process monitoring data. In an era of smart machines, now is the time for manufacturers to be bold and design greater contextual intelligence and machine learning to the board and circuit-level, so their machinery can deliver the insights needed to drive services revenue. The goal needs to be creating a more contextually intelligent series of machines. While OEE is not the main focus, it is one of several metrics indispensable in keeping machinery running and assuring maximum asset value.
Louis Columbus is currently serving as Principal of DELMIAWorks.
