machine icons

Perspectives

Predictive maintenance and the smart factory

Connecting machines to reliability professionals

Traditionally, maintenance professionals have combined quantitative and qualitative techniques to predict impending failures and mitigate downtime in their manufacturing facilities. Predictive maintenance (PdM) offers the potential to optimize maintenance tasks in real time, maximizing the useful life of equipment while avoiding disruption to operations.

Introduction

One important objective of any maintenance organization is to maximize asset availability. In this article, we focus on the applicability of PdM to fixed assets in manufacturing or warehouse automation. This could apply to many types of facilities, including a manufacturing shop floor, a warehouse or sortation facility, or assets used in mining or medical environments.

Poor maintenance strategies can reduce a facility’s overall productive capacity by 5% to 20%.1 Recent studies also show that unplanned downtime is costing industrial manufacturers an estimated $50 billion each year.2 This begs the question, “How often should an asset be taken offline to be serviced?” Traditionally, this dilemma forced most maintenance organizations to choose between two options: Maximizing the run time of an asset at the risk of asset downtime (run-to-failure) or maximize reliability of an asset through early replacement of potentially good components (time-based preventive maintenance).

The run-to-failure approach of maintenance can lead to catastrophic asset damage as parts begin to vibrate, overheat, and break, reinforcing the adage “pay me now or pay me later.” Additionally, while run-to-failure may be an acceptable approach for some assets, the unplanned downtime when the assets ultimately fail is almost always more expensive and time consuming. Conversely, a preventive maintenance strategy to frequently replace parts and service equipment can increase replacement costs over time and increase planned downtime and disruption to operations.

Predictive maintenance promises to provide the best of both worlds by aiming to reduce unnecessary preventive maintenance while ensuring that assets don’t face catastrophic failure.


Leveraging smart factory power

PdM aims to empower companies to maximize the useful life of their parts while avoiding unplanned downtime and minimizing planned downtime. With the advent of Industry 4.0 for manufacturing, companies can leverage technologies to monitor and gain deeper insight into their operations in real time to increase production efficiency and reduce costs. Simply put, a smart facility is one equipped with technology that enables asset-to-asset (i.e., machine to machine, or M2M) and asset-to-human (i.e., machine to human, or M2H) communication in tandem with analytical and cognitive technologies so that decisions related to the facility are data driven and on time.

PdM as a concept has been around for years and is based on the utilization of data from various sources such as critical equipment sensors, programmable logic controllers (PLCs), smart electronic devices, enterprise resource planning (ERP) systems, computerized maintenance management systems (CMMS), and manufacturing execution systems (MES). Smart facility management systems couple this data with advanced prediction models and analytical tools to predict failures and proactively address them. By contrast, traditional maintenance strategies to balance between run-to-failure or preventive maintenance often require time-consuming, manual data crunching and analysis to gain insights from the data being collected. While many organizations have had some success with these strategies, they typically rely heavily on “tribal knowledge” estimates or require in-depth knowledge of the equipment on an ongoing basis to stay accurate.

Moving toward PdM

Across industries, maintenance organizations are, by design or default, at different stages of maturity. Some may be running scheduled maintenance checks based on estimates or original equipment manufacturer (OEM) recommendations, while others may use statistics-based programs tailored to each asset. Still others are employing continuous monitoring technologies but are monitoring signals in isolation (univariant analysis), rather than leveraging predictive models.

There are steps to take on the journey toward reliability optimization, beginning with some of the basics of preventive maintenance and reliability-centered maintenance, while piloting PdM with one or two appropriate assets or facilities. Assets chosen for at least one of these pilots should be integral to operations and have enough run time to create baseline predictive algorithms.

Many of the technologies that make up a smart facility aren’t new but have become much more affordable, robust, and easy to integrate with big data platforms. Computing, storage, and network bandwidth are now available at fractions of the cost compared to 20 years ago,4 making piloting and scaling financially feasible. These are some of the technologies included in a smart facility that make PdM possible.

The Internet of Things (IoT)

IoT devices utilize internet infrastructure to stream continuous data, typically created in a continuous stream, from assets to private enterprise servers. IoT translates physical actions from assets into digital signals using sensors such as temperature, vibration, ultrasound, or conductivity. Data can also be streamed from other sources, such as an asset’s PLC, MES terminals, CMMS, or even an ERP system. IoT completes the first half of the physical-digital-physical (PDP) loop (figure 2). With the affordability of bandwidth and storage, massive amounts of data can be transmitted to give not only a full picture of assets in a single plant/facility but one of an entire production network as well.

Analytics and visualization

The second step in the PDP loop is to analyze and visualize digital signals using advanced analytics, predictive algorithms, and business intelligence (BI) tools. Many analytics platforms can incorporate unstructured data, cognitive technologies, asset learning, and visualization. Operations analysts, who are in touch with the manufacturing processes, can create dashboards using modern application program interfaces (APIs) created specifically for the everyday user.

Another trend is data moving back to the edge. Like the lean technique of storing tooling at the point-of-use, data computation is done at the “edge,” meaning it’s processed at the asset where it’s generated. Insights can be pushed directly to asset operators and maintenance technicians. In post-edge processing, this data can be pushed to outer nodes or a data warehouse on cloud to alleviate the burden on edge and core network traffic and to improve application performance.5

Closing the loop

After the signals have been processed, analyzed, and visualized, it’s time to turn those insights back into physical action. In some cases, the digital conclusions drawn may instruct robots or assets to alter their functions. In other cases, maintenance alerts will spur a technician into action. Consider a situation in which the predictive algorithms would trigger the creation of a maintenance work order in the company’s CMMS, check the ERP system for spares on hand, and automatically create a purchase request for any additional parts required. The maintenance manager would only need to approve the items in the workflow and dispatch the appropriate technician—all prior to machine failure.

A well-designed PdM solution can require orchestrated integration across an ecosystem of technologies as depicted in figure 3.

Potential benefits

At first glance, undertaking a PdM program might seem overwhelming. However, the benefits of digital transformation far outweigh the effort. These benefits can include:7

  • 5%–10% material cost savings (operations and maintenance, repair, and operations [MRO] material spend)
  • 5%–20% reduced inventory carrying costs
  • 10%–20% increased equipment uptime and availability
  • 20%–50% reduced maintenance planning time
  • 5%–10% reduced overall maintenance costs
  • Improved health, safety, and environment (HSE) compliance
  • Less time spent on brute-force information extraction and validation
  • More time spent on data-driven problem-solving
  • Clear linkages to initiatives, performance, and accountability
  • More confidence in data and information leading to ownership of decisions

 

Key adoption challenges

With the development of IoT technologies, reduced cost of data storage/computing, and advancements in artificial intelligence/machine learning (AI/ML) capabilities, industrial automation is growing rapidly. Yet, based on our experience, maintenance organizations have not been able to harness the power of these technologies beyond pilots for reasons described in figure 4.

 

A workable approach

Moving away from the pilot approach to a scalable and transformational approach requires strategic integration. With the help of an integrator, organizations can better leverage expertise to strategize, plan, and implement a holistic PdM program.

For example, Deloitte’s proven approach to adoption shown in figure 5 starts with vision and alignment of PdM goals with key stakeholders and focuses on identifying and prioritizing relevant use cases. Other best practices include selecting the right technology stack and ecosystem, a robust benefits case to drive action, and defining change management—all supported by a robust business case (cost vs. benefit) and a value tracking mechanism.

 

Learn more

Smart facilities using PdM are the future, but it can be difficult to determine the right investment path to value. Download this overview to learn more about Deloitte’s approach and offerings, and reach out to learn more. We can meet you wherever you are on your digital transformation journey.

Deloitte’s Experience

Deloitte’s approach to Predictive Maintenance is grounded in 15+ projects in the Smart Factory and Predictive Analytics domains

The Smart Factory @ Wichita

Seeing is believing for smart factory solutions. Our experiential facility, The Smart Factory @ Wichita, demonstrates advanced manufacturing techniques on a shop floor and shows how you can scale your endeavors across your enterprise. The net-zero, 60,000-square-foot space features an end-to-end smart production line, space for innovating with smart ecosystem sponsors, and experiential labs.

Learn more

Endnotes

1 Gary Wollenhaupt, “IoT slashed downtime with predictive maintenance,” PTC, accessed March 7, 2017.
2 IndustryWeek, “Unlocking performance,” Emerson, accessed September 26, 2022.
3 “Smart factory” broadly refers to a “smart” facility such as a manufacturing shop floor, warehouse, sortation facility, or intelligent building.
4 Brenna Sniderman, Monika Mahto, and Mark Cotteleer, Industry 4.0 and manufacturing ecosystems: Exploring the world of connected enterprises, Deloitte University Press, February 23, 2016.
5 Adam Mussomeli, Doug Gish, and Stephen Laaper, The rise of the digital supply network: Industry 4.0 enables the digital transformation of supply chains, Deloitte University Press, 2016.

Fullwidth SCC. Do not delete! This box/component contains JavaScript that is needed on this page. This message will not be visible when page is activated.

Did you find this useful?