Using Predictive Maintenance to Optimize Thermal Processing Operations

Wouldn’t it be nice if your furnace was able to let you know when it wasn’t operating properly? Or when it is necessary to rebuild a vacuum pump? Perhaps it could tell you if the next cycle risks discoloration? Or that the leak back test, scheduled for three weeks’ time, won’t be passed?

Imagine if your furnace was able to warn you about the failure of a heating element, order the new part required, and schedule the necessary service for its installation?

These hypotheticals are the motivating influences driving predictive maintenance technology to the forefront of maintenance strategies and product development for industries all over the world. In the coming years, customers will expect every heat treatment furnace to have the ability to leverage the Internet of Things in order to perform such analyses (Fig 1).

This chart displays the progression of equipment as it becomes increasingly integrated with the Internet of Things. "Service" represents the capabilities of a traditional heat treatment furnace, while "Analyze" represents a heat treatment furnace integrated with predictive maintenance (courtesy of PTC).

Figure 1. This chart displays the progression of equipment as it becomes increasingly integrated with the Internet of Things. "Service" represents the capabilities of a traditional heat treatment furnace, while "Analyze" represents a heat treatment furnace integrated with predictive maintenance (courtesy of PTC).

In the thermal processing industry at present, there is a clear result if a heat treatment furnace breaks. Production stops immediately, and the engineers required to fix the problem may not be on site at the time.

Consequently, companies must account for unscheduled downtime until the problem can be fixed, as well as possibly paying overtime wages for the required workers, the costs of rushing critical part shipments, and other such negative consequences.

As a means of tackling this issue, Ipsen's PdMetrics® software platform is specifically designed for predictive maintenance, the ultimate goal of which is the undertaking of maintenance at scheduled times when this maintenance is the most-cost effective and prior to the equipment’s performance levels dropping below a pre-set threshold.

Clearly, predicting maintenance requirements in advance and preventing failures prior to their occurrence offers numerous advantages. These include improving furnace reliability and uptime, avoiding high-cost events, reducing the need for regular maintenance or repairs, and minimizing shop personnel’s maintenance burdens resultant upon unanticipated downtime.

Additionally, a smart, connected furnace - which is able to monitor in-service equipment in order to capture data which is of use in refining furnace operations and reporting when service is required - is the result of the integration of predictive maintenance.

Predictive maintenance software can identify maintenance trends, deteriorating conditions, and more, by analyzing the critical furnace data. This allows users to plan ahead, either by scheduling the required personnel in order to preform maintenance, or by making sure the required furnace parts are in stock.

Predictive maintenance also allows furnace manufacturers to gather data concerning the overall performance of the furnace. This helps with future innovations and the continuous improvement of products.

In order to maintain equipment in the most effective manner, the best maintenance programs will offer a mix of corrective, preventative, and predictive maintenance overall.

The addition of predictive maintenance, however, is what helps to make a cost-effective program which can identify issues prior to their occurrence, schedule the required maintenance, and properly prepare for any corrective maintenance that might be necessary.

Predictive maintenance will continue to be used more widely as modern equipment evolves towards the goal of optimizing operations and reducing unplanned downtime.

Companies can get a better understanding of how to successfully increase production, optimize operations, and reduce both unplanned downtime and unnecessary costs by analyzing the growth of predictive maintenance as a tool for assessing maintenance and performance needs. The same can be said of an examination of how Ipsen’s PdMetrics software platform is revolutionizing the thermal processing industry.

PdMetrics® Software Platform – Powerful but Simple Diagnostics

Control algorithms and proprietary sensors are utilized by Ipsen's PdMetrics® software platform in order to determine the appropriate timing for the replacement of parts and servicing of the furnace. Both Ipsen and the customer receive an automatic alert when the software determines an action is necessary.¹ This alerts the customer in advance and allows necessary parts to be delivered, as well as enabling service personnel to be dispatched on a scheduled date. Other essential features include:

  • Stand-alone interface screen which can integrate with numerous furnaces
  • Notifications for the replacement of furnace parts based on wear and usage characteristics
  • Users can receive alerts by text or email, courtesy of smart connectivity
  • Ipsen's Service Cloud offers technical support which enables Ipsen experts to rapidly access the relevant data and provide essential support¹

Ultimately, by monitoring critical data and key parameters such as pressure, temperature, and vibrations via the PdMetrics software platform, users can improve the integrity and health of numerous systems: the pumping system, the hot zone, the cooling system, and vacuum integrity (Fig. 2).

The PdMetrics® software platform dashboard, where users can monitor the health and integrity of the hot zone, pumping system, cooling system and vacuum integrity.

Figure 2. The PdMetrics® software platform dashboard, where users can monitor the health and integrity of the hot zone, pumping system, cooling system and vacuum integrity.


The Maintenance Spectrum

Yet, to acquire a complete understanding of predictive maintenance’s emergence and the concomitant benefits, two common maintenance forms which are currently in use must first be assessed: corrective and preventative maintenance.

Each of these maintenance strategies has its own requirements and is frequently performed together with another form of maintenance in order to maintain production levels, as well as control, or reduce, maintenance costs.

Corrective Maintenance

Corrective maintenance is considered to be one of the most basic maintenance approaches and is generally used following a furnace’s failure to function correctly. Consequently, corrective maintenance frequently requires a look at the furnace’s common symptoms, in order to ascertain the likely causes of failure and, ultimately, fix them [1].

There are a number of disadvantages inherent in performing corrective maintenance. The main disadvantage is that failure is permitted to occur in the first instance. Consequently, disruption to production cannot be completely avoided. It also becomes increasingly expensive and difficult to reduce the time needed to fix the issue, as users need to be constantly prepared for things to go awry

As a result, a vast range of spare parts must be kept close by all the time, which necessitates a significant amount of storage space. Maintenance personnel must also be paid overtime wages for correcting the issue, and there are further issues, all of which can reduce profit levels.

Preventative Maintenance

On the other hand, preventative maintenance requires frequent inspections and maintenance of the furnace equipment prior to the occurrence of any major failings.

Preventative maintenance (PM) programs typically include equipment servicing, regular inspections, repairs, and replacements, and constitute an essential aspect of protecting one’s equipment. Companies can schedule downtime in advance using PM programs, which help to provide predictable, yearly maintenance costs [2].

In contrast to corrective maintenance which waits for failures to occur prior to fixing them, preventative maintenance involves regular prevention of errors prior to their occurrence, in order to keep the furnace in fantastic working condition.

Consequently, preventative maintenance is frequently seen as an effective way of preventing unplanned downtime and protecting equipment. The two sub-categories of preventative maintenance are time-based and condition-based maintenance.

Time-Based Maintenance

Time-based maintenance works by implementing a preventative action in response to a specific trigger. Triggers can be based on several different items, including production quantities, lead times, the amount of time passed, and others.

A frequently used form of time-based maintenance is the changing of oil in a car, as this action is performed either at fixed intervals of time or according to a specific mileage.

However, there are disadvantages inherent in this form of maintenance. For instance, as parts with low wear are exchanged in accordance with this time-based principle, they must be replaced more often. Consequently, large numbers of low wear parts must be purchased and stored as a means of preventing unnecessary downtime, thereby increasing costs.

Ultimately, if the expenses created by time-based maintenance are lower than the costs which would occur due to unplanned downtime scenarios or other maintenance forms (such as corrective maintenance), it is worth utilizing [3].

Condition-Based Maintenance

Condition-based maintenance is similar to time-based maintenance, as it is also reliant upon a trigger to indicate when preventative action is necessary.

In this instance, the desired condition of the equipment acts as the trigger. As they are regularly inspected at time-based intervals, wear-related conditions are recorded either via human inspection or using sensors. They are subsequently compared against the required wear which is allowed for safe operation. Maintenance of the necessary systems and/or parts needs to be undertaken if that value is exceeded [4].

The replacement of car tires once the minimum tread depth has fallen below a pre-set limit represents one example of condition-based maintenance. In the context of working with a heat treatment furnace, another example is adjustments to the pump water miser when the operating temperature drops below the recommended parameters.

However, there are certain limitations inherent in this form of maintenance. One limitation is that when a limit is exceeded, maintenance needs to be performed immediately. This results in maintenance periods which are unpredictable, and necessitates keeping a large number of spare parts in stock all the time. Both of these limitations increase costs of storage and personnel.

Condition-based maintenance is similar to time-based maintenance in that it is only worth utilizing when the overall benefits outweigh the overall expenses, as well as the opportunity cost of using other maintenance forms.

Predictive Maintenance

The idea behind predictive maintenance is the application of analytics in order to identify risks of failure, thereby helping to prevent failures before they occur. The optimum predictive maintenance programs are frequently used together with a preventative maintenance program, and are usually designed in order to prevent failures of equipment and schedule maintenance in accordance with requirements.

Predictive maintenance delivers essential data by monitoring the furnace, its performance, and other parameters. This data can subsequently be analyzed in order to ascertain when maintenance ought, or needs, to be performed. This delivers a heat treatment furnace which is able to perform the hypotheticals discussed previously.

Furthermore, numerous aspects including the reliability, efficiency, and performance of the furnace, can be optimized by this data. Consequently, in several ways, predictive maintenance is an enhancement of the condition-based preventative maintenance system.

Predictive maintenance, in opposition to the aforementioned maintenance forms, enables the planning of available resources to be total, thereby helping to minimize the costs of unnecessary spare parts, storage, and personnel. Additionally, it is able to effectively identify problems occurring between scheduled inspections.

Ultimately, there is typically a high benefit-to-cost ratio associated with predictive maintenance systems, which leads to specialized solutions for the key components of a larger system, such as a heat treatment furnace.

The Convergence of the Internet of Things and Big Data

One of the essential factors influencing the numerous decisions and actions taken by a range of businesses is improving operation. The International Data Corporation (IDC) recently demonstrated this fact in a survey of businesses and operations professions, which concluded, "the leading driver of Big Data and analytics projects is 'product or service improvement and innovation'" [5].

In order to comprehend these projects’ increasing development, it is necessary initially to understand the convergence of Big Data and the Internet of Things.

Understanding the Internet of Things

Coined in 1999 by the co-founder and former executive director of the Auto-ID Center, Kevin Ashton, the term ‘the Internet of Things’ (IoT) was a way of describing the manner in which physical objects are connected to the internet:

"Today computers – and, therefore, the Internet – are almost wholly dependent on human beings for information. Nearly all of the roughly 50 petabytes (a petabyte is 1,023 terabytes) of data available on the Internet were first captured and created by human beings – by typing, pressing a record button, taking a digital picture or scanning a bar code [...] The problem is, people have limited time, attention and accuracy – all of which means they are not very good at capturing data about things in the real world [...] If we had computers that knew everything there was to know about things – using data they gathered without any help from us – we would be able to track and count everything, and greatly reduce waste, loss and cost. We would know when things need replacing, repairing or recalling, and whether they were fresh or past their best" [6].

Even though the concept of the IoT dates back to 1999, it was only in 2010 that it began to gain traction. Overall, the IoT includes connections which go beyond the industrial context, including wearable devices on people (i.e., physical objects which are linked via wireless and wired networks). One such example is a Fitbit², which monitors various aspects of a person’s health and records that data using a network connection.

Defining Big Data

If the IoT is considered as a vast array of Internet-connected sensors which are attached to a variety of ‘things’, Big Data is a term denoting the enormous volumes of data generated by these ‘things’.

A smartphone can be taken into consideration to demonstrate this. When people carry these phones around, many of their regular activities (both virtual and physical) may be tracked, analyzed, and acted upon. The data created by people’s smartphone activity is considered to be ‘Big Data’.

The four Vs (veracity, variety, volume, and velocity) characterize Big Data. Put simply, "connected, sensor-instrumented industrial devices are yielding data" rapidly (velocity), in large amounts (volume), in a mixture of "structured, semistructured or unstructured" information (variety) and this data is often "noisy and of uneven quality" (veracity), meaning certain data may be more accurate than others depending on its source [7].

Analytics initiatives, such as predictive maintenance, then use this data to "[discover] early signals that predict machine failure to determine priorities for asset maintenance or anticipating a shift in demand that will impact operations delivery capability" [8]. This combines to contribute to improving operations via the utilization of Big Data and the Iot in predictive maintenance programs.

Development of PdMetrics Software Platform

In order to provide a means for companies to create value from the vast array of data which is available from equipment and processes run in a furnace, Ipsen developed the PdMetrics software platform. This provides businesses with a means of reducing unnecessary downtime in an efficient and cost-effective manner, whilst also optimizing operations.

Furnace users are provided with four key features by the PdMetrics software platform. These are an intelligence maintenance routine, continuous optimization of furnace usage, a Diagnostic Helper with access to essential resources and tools, and the capability of achieving maximum equipment performance. All of these features interact with one another in order to deliver an integrated user experience.

The capability of monitoring critical data and essential furnace parameters has become a fundamental requirement for companies focusing on reducing machine failures and, consequently, maintenance costs.

Studies in the industry have indicated that, "the failure of a machine critical to an operational process can have a major impact on the revenue that can be generated by a firm [...] Accurate, timely predictions can save millions of dollars in maintenance costs" [9].

Ultimately, the implementation of a software platform for predictive maintenance offers solutions for all levels of personnel. Furnace operators are now easily able to monitor the health and ongoing status of the furnace, while managers are given a total overview of the operation in its entirety, courtesy of a network of connected furnaces with a greater combined intelligence (Fig. 3).

The above chart illustrates the degree to which having a connected furnace positively impacts different departments and aspects of a company

Figure 3. The above chart illustrates the degree to which having a connected furnace positively impacts different departments and aspects of a company's operations – from the furnace operator to management.

Below, the ways in which the four main features operate when monitoring the hot zone’s necessary critical data are taken into account:

Maximum Equipment Performance

In order to guarantee that the hot zone performs to a maximum, factors such as hot zone cleanliness, and resistance to ground, heat loss, and open circuits are monitored. This consequently helps to avoid arcing, which has the potential to cause damage to the heating elements; prevents high-energy consumption by the hot zone; and makes sure of proper heating uniformity.

Diagnostic Helper

In order to help furnace users improve their ability to determine problematic symptoms’ causes, an Ask the Expert feature and leak diagnostic procedures are offered by the Diagnostic Helper. Numerous valuable resources are also offered, such as tips on valve troubleshooting sequences, maintenance troubleshooting, and more.

For instance, if a furnace is experiencing trouble attaining vacuum levels (i.e., evacuating), it may indicate the presence of a leak. In this scenario, users are able to use the Diagnostic Helper’s resources in order to ascertain the root cause before the hot zone is slowly degraded by the leak.

Alternatively, if outgassing, dirty parts, and/or lengthy pumpdowns suddenly begin to be experienced by users, they can use the weather station to determine whether the levels of humidity are the cause, or whether these symptoms may indicate a more substantial problem.

Intelligent Maintenance Routine

Although having a set preventative maintenance (PM) program as well as corrective maintenance capabilities remains essential, existing maintenance and PM programs are built upon by the PdMetrics software platform.

This platform incorporates an intelligent maintenance routine which provides an automatic maintenance reminder based on component usage and furnace performance. Consequently, instead of checking for discoloration or replacing hot zone components based upon the amount of time passed, users are given notifications when the hot zone indicates a need for maintenance.

Resources for maintenance personnel, replacement parts, and more can thus be allocated according to a schedule. Additionally, if an action is required and/or an anomaly or error is experienced by the furnace, the software platform maintains a log of all the issues or errors the furnace has ever encountered.

Completing this software platform’s ability to deliver an intelligent maintenance routine is its incorporation of calibration due dates. The software is aware of the times when calibration is required, providing sufficient time to plan in advance in order to minimize the interruption to production.

Optimize Furnace Usage

Lastly, users can experience increased operational visibility, as well as improving their furnace usage. The software platform can integrate with a variety of furnaces and has a smart connectivity capability which enables users to receive alerts by text and/or email. This has made it even simpler both to monitor several furnaces in the same location (Fig. 4), as well those in different facilities. Factories are consequently connected globally in a manner never previously possible.

Through analysis of the gathered data, the ability of users to implement any required adjustments, which will help to improve or refine the equipment’s performance, is enhanced. Users are also better able to delegate resources such as parts and personnel, according to overall requirements.

The PdMetrics® software platform dashboard allows users to monitor the health and performance of multiple furnaces at once. Whenever an action is required or an error is experienced by one of the furnaces, it is signified by a yellow caution or a red danger warning sign

Figure 4. The PdMetrics® software platform dashboard allows users to monitor the health and performance of multiple furnaces at once. Whenever an action is required or an error is experienced by one of the furnaces, it is signified by a yellow caution or a red danger warning sign.


What if a furnace was able to inform users when it isn't operating correctly?

This simple, yet important question is the impetus behind the continuous evolution of maintenance procedures and methods. This continuous evolution is vital if production is to be successfully increased and operations optimized, while manufacturing costs and unplanned downtime are simultaneously decreased. Predictive maintenance constitutes one answer to this what if, and it is the next step on the maintenance evolution path.

As this path unfolds, several companies are beginning to learn that, "predictive maintenance analytical models can guide managers to better decisions on how to deploy assets and when to maintain them to ensure safe, efficient, and optimized operations" [10].

Ipsen has discovered that the PdMetrics software platform for predictive maintenance delivers a new, innovative solution for the thermal processing industry, courtesy of its real-time monitoring of critical systems and its ability to guarantee ultimate equipment performance.

The benefits of this software include integrating the software platform with the current service department, in order to improve the tracking and scheduling of regular preventative maintenance activities; predicting and scheduling service based on the operational history of the furnace; and reducing unplanned downtime, courtesy of the capability of determining inventory needs in advance, and correcting problem areas before they become critical.

Ultimately, the PdMetrics software platform represents an important example of the emergence of predictive maintenance as a tool for analyzing maintenance needs and equipment performance, as well as how Big Data and the Internet of Things have impacted upon the world of heat treatment.


[1] Grann, Jim, "Protecting Your Vacuum Furnace with Maintenance," Ipsen White Paper, Industrial Heating (2015),

[2] Ibid.

[3] Ahmad, Rosmaini, and Shahrul Kamaruddin, "An overview of time-based and condition-based maintenance in industrial application." Computers & Industrial Engineering, Vol. 63, No. 1 (2012), pp. 135-149.

[4] Ibid.

[5] Morris, Henry D. et al., "A Software Platform for Operational Technology Innovation," International Data Corporation (2014), pp. 1-17.

[6] Ashton, Kevin, "That 'Internet of Things' Thing," RFID Journal, Vol. 22, No. 7 (1999), pp. 97-114.

[7] Morris, pp. 7.

[8] Ibid., 1.

[9] Ibid., 4.

[10] Ibid., 16.

This information has been sourced, reviewed and adapted from materials provided by Ipsen.

For more information on this source, please visit Ipsen.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Ipsen. (2019, August 20). Using Predictive Maintenance to Optimize Thermal Processing Operations. AZoM. Retrieved on January 23, 2020 from

  • MLA

    Ipsen. "Using Predictive Maintenance to Optimize Thermal Processing Operations". AZoM. 23 January 2020. <>.

  • Chicago

    Ipsen. "Using Predictive Maintenance to Optimize Thermal Processing Operations". AZoM. (accessed January 23, 2020).

  • Harvard

    Ipsen. 2019. Using Predictive Maintenance to Optimize Thermal Processing Operations. AZoM, viewed 23 January 2020,

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback