Navigation
Home
News
Newsletter
Knowledge Bases
Tutorials
Directory
Books
Maintenance-Tips
Conferences
Forums
Photo Albums
Reliability Radio
Benchmark
Calendar
Gadgets
Jobs
Network Links
Archives
en Español
XML/RSS Feeds Advertise

Condition-Based Maintenance Using Wireless Monitoring: Developments and Examples by Wayne Stargardt, Aleier, Inc. 

Summary 

New developments in wireless data collection technologies are lowering the costs of gathering the data necessary to perform predictive maintenance for a broader range of equipment. Wireless-based continuous, automatic monitoring by the CMMS can extend condition-based maintenance strategies to more equipment as well as improves the performance of corrective maintenance. This paper will review recent developments in wireless data acquisition technologies and describe typical costs in implementing them in practical situations. The paper will also describe how to integrate this data with a CMMS, and how the CMMS can take advantage of equipment condition data. Finally, the paper will profile some examples in which condition-based maintenance using wireless monitoring is improving operations.

Overview of Condition-Based Maintenance

Maintenance organizations employ several different maintenance strategies to keep their fixed assets and equipment operating reliably and at peak performance. One of the tenants of the discipline of Reliability Centered Maintenance is that organizations should employ an appropriate maintenance strategy for each asset, depending on its importance to the operation, the costs of maintaining it, and its failure modes and failure frequencies.

The original maintenance strategy is corrective or reactive maintenance, under which equipment is repaired or replaced only after it has failed or suffered serious performance degradation. This “run to failure” strategy is still largely appropriate for assets whose failure will not compromise operations, which can be returned to service quickly and easily, or whose failure modes and timing do not show a statistically significant pattern.

Because preventive maintenance is expensive, and because enterprises are increasingly pressured to reduce their costs, organizations are developing a new type of maintenance strategy. Under this strategy, the condition of the asset is monitored regularly until it begins to give evidence of deteriorating performance or incipient failure. Maintenance is then performed “just-in-time” to prevent asset failure.

The more common version of this new maintenance strategy, predictive maintenance, is also based on the statistical pattern of failure exhibited by equipment. Many common failure modes are presaged by changes in equipment behavior, and monitoring these parameters can provide early warning of failure or degradation. Under predictive maintenance, these parameters are monitored periodically, usually based on manually collected data. For example, predictive maintenance is often based on analysis of bearing heat signature, lubricant condition, and rotating equipment vibration.

Condition-based maintenance is similar to predictive maintenance in that analysis of the equipment condition is used to determine the timing for “just in time” maintenance. With condition-based maintenance, equipment operating conditions are monitored continuously and in real time to identify the need for maintenance. Continuous monitoring is obviously more timely than the manual, periodic data collection used for preventive maintenance, but it can also be less expensive, more reliable and more accurate.

Both predictive maintenance and condition-based maintenance can be less expensive than preventive maintenance while providing the same benefit of high asset availability and reliability. By its very nature, preventive maintenance means that maintenance is being performed more often than is necessary. Since maintenance consumes both labor and parts, this strategy has a measureable cost of “over-maintenance.” Additionally, preventive maintenance often requires that assets be taken off-line during servicing, incurring a cost to the organization for this downtime and lost capacity. Finally, more frequent maintenance involves more frequent intrusions into the equipment, which itself increases the chance of asset or system failures. Predictive and condition-based maintenance avoid these penalties of over-maintenance.

There are a number of factors that have limited the adoption of condition-based maintenance besides the fact that it is a relatively new strategy. Most of the equipment that could be subject to condition-based maintenance is not currently instrumented with sensors. Data on equipment condition cannot be collected without installing sensors on the equipment along with a means of collecting the sensor data. In those cases when equipment is already instrumented, it is done for control and automation and does not track the operating parameters that could identify incipient failure.

Even instrumented equipment is often not connected into a data collection network that allows real-time monitoring. These unconnected instruments or sensors usually provide local displays or store readings in data loggers, and data is collected from them manually. Implementing a condition-based maintenance program usually requires retrofitting additional sensors and a data collection network. This is a primary impediment to the broader adoption of condition-based maintenance.

Wireless Technologies for Maintenance Applications

The prospects for condition-based maintenance are improving rapidly. Condition-based maintenance is being enabled by several advances, but especially by advances in wireless networking technology. Just as people have become interconnected by the Internet over the past decade, so equipment and devices are beginning to become connected to the Internet as well. Much of the equipment that is being connected is already in service, so its connectivity needs to be retrofitted. A wide range of wireless technologies are being used to provide that connectivity.

Wireless technology has several advantages for retrofitting equipment for data collection. Increasingly, wireless technologies are less expensive than either retrofitting wired connections or collecting data manually. Wireless technologies provide the automatic and continuous connections provided by wires, and wireless is approaching wired connections in its reliability.

Previously, a number of concerns have kept wireless technologies from being used in the type of data collection needed for condition-based maintenance. These concerns are rooted in the understanding that wireless connections are not exactly equivalent to wired connections. The foremost issue is the perception that wireless connections are unreliable since they can be interrupted by excessive distance, signal blockage or interference from nearby RF emitters. Other concerns have been limited range of the wireless connections, as well as lower throughput and higher latency than wired connections. Security is often raised as an issue since wireless signals can be received across a broad area. Perhaps more important than these technical concerns, wireless connections have not been used more frequently because they have often cost more than simple wired connections. 

Continuing advances in wireless technology over the last decade have overcome most of the traditional concerns about using wireless-based connections. Improved radio designs have made significant advances in improving the range and reliability of wireless links. In addition, more sophisticated networking protocols have been incorporated into the designs of these radios to further improve the reliability and performance of the wireless networks. Most of these modern wireless networks also incorporate encryption, authentication processes and other security features to allay the concerns about eavesdropping and other security issues. Finally, the high manufacturing volumes provided by the cellular and Wi-Fi markets have driven costs down significantly so that wireless connections are often much less expensive than other alternatives. 

Wireless connectivity is poised for broad adoption in the industrial and commercial automation market similar to the situation with Ethernet 25 years ago. At that time, Ethernet was moving from defense and educational applications into broader commercial and office use. There was significant resistance to the use of Ethernet in industrial applications, however, since it was perceived that the shared nature of Ethernet did not provide exactly the same, deterministic reliability as direct wired connections. The increasing familiarity with Ethernet, together with the development of distributed control system architectures, has led to widespread adoption of Ethernet in industrial environments for several aspects of monitoring and control applications. Wireless connectivity is also starting to be adopted more broadly as its capabilities are better understood.

There are a number of wireless technologies that are available for continuously monitoring equipment condition. Cellular telephone networks are being used to monitor equipment, particularly in remote locations or when the equipment is frequently mobile (i.e., vehicles, construction equipment). Equipment is even being monitored over satellite communication networks for those areas beyond cellular coverage.

The major advances in using wireless for equipment monitoring, however, is using low cost, low-powered, unlicensed wireless technology. Most wireless communication requires each transmitter to be licensed by government regulators, including cellular telephones, radio and TV broadcasting, and microwave links. There are certain parts of the radio spectrum, however, in which any low power wireless device is allowed to operate with requiring a government license. The most familiar example is Wi-Fi, the wireless local area networking technology also referred to as 802.11. Originally developed for connecting computers, Wi-Fi can monitor equipment in situations where Wi-Fi networks already exist and where the radio can be connected to power. Related wireless technologies have been used to develop wireless connections specifically for industrial applications. These industrial wireless networks are usually proprietary and are available for a variety of frequencies and network configurations (i.e., point-to-point, point-to-multipoint.) The variety and specialization of these wireless industrial networks are similar to the variety of fieldbus wired networks for the same connectivity applications.

Wireless sensor networks are yet another connectivity alternative and are a relatively new development. These networks use some version of a mesh architecture, in which one radio’s data is relayed on to the final destination by other radios, and usually by several. These networks incorporate sophisticated intelligence that allow them to configure themselves automatically, and to develop the routing schemes so that the fewest number of radios are involved in conveying the data while also routing around interference and obstacles. These wireless sensor networks provide broader aggregate coverage than simpler point-to-point wireless technologies, while also achieving more reliable operation. Wireless sensor networks greatly expand the number of monitoring and control applications that can be served by wireless connections.

The challenge of using wireless connections for monitoring equipment condition is that no single one of the different wireless technologies is the most cost effective solution for all applications, so each application must be analyzed to select the appropriate wireless technology. In general, the broader the area in which the monitored equipment is located, the more expensive is the wireless technology that must be employed. Also, the distance between the individual pieces of equipment will determine whether a wireless sensor network can be used cost effectively, or whether another wireless technology is needed. Different wireless networks are also optimized for the volume of data to be transmitted, although moving larger volumes of data generally requires a more expensive solution. Some wireless technologies operate on frequencies that provide better coverage in buildings, but these frequencies are only available in certain countries. Some wireless networks enable their radios to operate for extended periods on battery power, while other wireless networks cannot be used practically without connection to a power source. Choosing the appropriate wireless technology for monitoring equipment involves making a number of choices and tradeoffs.

The cost of wireless connections is an important consideration in using this technology, although costs have improved dramatically in recent years. For industrial wireless networks that only communicate over relatively short distances, including wireless sensor networks, the cost of a complete radio has generally dropped to under $300, and sometimes under $100. Wireless industrial radios that communicate over somewhat longer distances can cost between $300 and $800. For communicating over distances measured in miles, radios can be bought for $500 to $1000. These wireless costs compete favorably against retrofitting wiring, in which the total cost can approach $100 per foot of wiring installed.

Wireless Standards Enable Adoption

Just as wireless local area networking became more attractive and cost effective after the development of the IEEE 802.11 standards (also referred to as Wi-Fi), emerging standards in wireless sensor networks will make them more acceptable and attractive. There are three major standards initiatives that are guiding the maturation of wireless sensor networks: ZigBee™, Wireless HART and ISA 100.

ZigBee is the more mature of the standards in that its first version has been released for several years and there are several manufacturers that provide ZigBee-based products. ZigBee is a general worldwide standard for low-power wireless mesh networks that is defined by an open standards organization called the ZigBee Alliance. The ZigBee standard has been designed to serve a wide range of applications from industrial automation to building automation to advanced metering to home automation. The most current implemented version of the ZigBee standard is the second major release, called ZigBee 2006. There are already multiple manufacturers shipping chip level implementations of ZigBee 2006, and a number of products incorporating these chips are also currently shipping. The next version of the ZigBee standard, ZigBee 2007, continues to add more functionality and has already been ratified. Chip level products that implement ZigBee 2007 are just starting to be announced. ZigBee is a well-established standard that has been adopted by a number of manufacturers and is continuing to evolve.

Wireless HART is the counterpart to the HART intelligent sensor standard that has been in existence for wired industrial automation and control applications for over a decade. Wireless HART is targeted specifically at monitoring and open loop control applications in industrial environments. Wireless HART is part of the overall HART Version 7.0 specification that was released in the fall of 2007. Products using the Wireless HART protocol have been announced and are expected to begin shipping in 2008. Products using Wireless HART should be especially useful for retrofitting condition-based monitoring for maintenance purposes.

The third important wireless sensor networking standard is ISA 100, which is being developed by the Instrumentation, Systems and Automation Society (ISA). The first iteration of this standard, ISA 100.11a, is still under development, but it should be ratified in 2008. ISA 100.11a targeted at monitoring and open loop control applications in process manufacturing industries, such as oil refineries, power plants and paper mills. Products employing this first version of the standard should ship in 2009. ISA is continuing to develop future versions of this standard for discrete manufacturing and for asset identification and management, and they are also working to enable interoperation with Wireless HART devices. ISA has developed a number of important standards for industrial automation, and ISA 100 will probably have a significant impact on the industry.

Wireless technology, including the development of industry standards, has evolved over the past few years so that it is a practical and affordable option for connecting equipment to implement condition-based maintenance.

Implementing Condition-Based Maintenance Management

Most larger organizations use an automated, computer-based system for managing maintenance operations. The primary function of these systems is to track and process maintenance work orders. 

Maintenance work orders are generated in two major ways – manually or automatically. Many work orders are created by manually entering requests for corrective or reactive maintenance work to fix immediate problems. On the other hand, most preventive maintenance work orders are generated automatically, usually based on a schedule of preventive maintenance for each piece of equipment. The maintenance management system will often modify the preventive maintenance schedule based on the age of the equipment, its recorded condition, and its corrective maintenance history.

Most predictive maintenance work orders are also created manually. Under most predictive maintenance methods, equipment condition or health is analyzed by an independent system using sampled data to determine whether preventive maintenance is required. If preventive maintenance is indicated, the work orders for that activity are created manually in the maintenance management system just as with corrective maintenance work. Using this approach, the computerized maintenance management system (CMMS) itself does little different when implementing a corrective maintenance strategy or a predictive maintenance strategy for an asset.

The maintenance management system or CMMS is much more involved in implementing condition-based maintenance strategies for assets because much more of the operational process is automated. Readings from sensors and instruments on the monitored equipment are communicated to the CMMS continuously, or at least frequently enough to achieve the same result. This data is analyzed automatically by the CMMS to evaluate the condition of the asset. This analysis is performed using rules or algorithms programmed into the CMMS based on the known failure modes and their early warning indicators for each individual piece of equipment. When the assessment of asset condition indicates a need for maintenance, the CMMS automatically generates a work order for that maintenance. Maintenance professionals do not need to become involved in the process until the maintenance work order is created.

Condition-based maintenance takes advantage of the automated management capabilities of a CMMS. It uses the equipment list that already exists in the CMMS, together with detailed information on each asset and the maintenance strategy selected for that asset. Incoming data can be automatically associated with the correct asset, analyzed and stored for future review or display. The CMMS can analyze the condition data using any number of simple or complex rules, immediately and consistently, to assess the underlying health of the asset. The CMMS can perform the tedious task of monitoring machine health to detect the exception conditions that indicate the need for maintenance, and it performs that task more reliably and less expensively than any other approach.

Example Applications

Some practical, operating examples can illustrate the different ways in which condition-based maintenance can be performed and how wireless technology can be used to implement it.

One example is the InVision™ Downtime Reduction System from the Bussmann division of Cooper Industries. Bussmann is the 90 year old subsidiary of a Fortune 500 company that is the largest manufacturer of industrial fuses in the world.  Cooper Bussmann fuses protect machinery and people in a wide range of industrial and commercial applications. These facilities blow fuses every day. While the fuse prevents damage from electrical surges and short circuits, Cooper Bussmann determined that such an open circuit event results in 41 minutes of downtime on average. The downtime causes a loss of production, supply chain interruption, and idling of the workforce, and it can cost a company from $125,000 to $750,000 per event. Cooper Bussmann developed InVision to help reduce the downtime caused by blown fuses.

The InVision system reduces downtime by accelerating the maintenance organization’s response to a blown fuse. The InVision system involves installing small, low cost, battery-powered wireless sensors on each fuse and circuit breaker throughout the plant or facility. The sensor detects when a fuse blows, or a circuit breaker trips, and it broadcasts that alert. The alert is communicated through the facility over a wireless sensor network infrastructure to a gateway connected to the Internet. The alert is then transmitted to the Cooper InVision Command Center. The Command Center automatically notifies the maintenance organization for that plant that a circuit event has occurred, the specific location of the fuse or circuit breaker, the model number of the replacement fuse, and any special instructions required to replace the fuse. Cooper Bussmann estimates that this automated alert and notification process reduces the mean time to restore service by an average 65%, which substantially reduces the cost of downtime.

Cooper’s InVision system uses a robust wireless mesh network for several reasons. Wireless systems are ideal for retrofitting existing facilities since most fuse and circuit panels are not previously instrumented nor connected, and it is less expensive to add connectivity through low power wireless links. It is also easier to reach difficult locations by using wireless communications. Finally, wireless sensor networks deliver the level of reliability required for industrial applications using a number of technological innovations.

Another example of using wireless technology to enable condition-based maintenance is a connection that RF Monolithics installed for a Department of Defense customer. This customer manages its assets and maintenance operations using an Enterprise Asset Management (EAM) system from Aleier, Inc., a subsidiary of RF Monolithics. Enterprise Asset Management systems usually include all of the standard functionality of a CMMS, and Aleier’s system is no exception. Some of the important assets in the customer’s facilities are large, walk-in freezers that store certain expensive, mission-critical materials that must be kept frozen. Because of the critical nature of the materials, any temperature alarms generated by the freezers’ stand-alone control systems require an immediate response by the maintenance staff without waiting for a work order. Unfortunately for the maintenance organization, the built-in temperature monitoring system on the freezers generated a high number of false alarms that are triggered during normal defrost cycles and by the weekly switchover of the redundant refrigeration compressors. This created significant wasted time by maintenance technicians that the customer wanted to eliminate.

This problem was addressed by installing a separate temperature monitoring system, which was provided by Cirronet, another division of RF Monolithics, that provides industrial wireless communications systems. In the Cirronet solution, separate temperature sensors were installed in the freezers to measure temperature independently. Because an Internet connection could not be provided to this building on an active military base, Cirronet communicated the temperature readings by installing a direct radio connection between the storage facility and the maintenance building over 2 miles away. The temperature readings were then communicated over the Internet to the centralized Aleier EAM system.

Figure 1
Example Monitored Freezer Temperature

This solution produced several benefits. First, false alarms were reduced by relying on the temperature readings analyzed by the EAM system, in which a more sophisticated rule looks at the relative trend of the temperature readings and the length of time that the temperature is above a monitoring threshold before triggering an alert (Figure 1). Second, for security, the EAM can also track whether the temperature crosses a critical threshold level and independently trigger an alarm if it does. Third, maintenance technicians can view the recent temperature profile and trend from within the Aleier EAM using any Internet connected PC. And finally, high temperature alarms are created within the EAM system, so that maintenance events are handled and managed consistently, and maintenance personnel are all scheduled and managed through one system and process. 

Energy management is another example in which wireless communications can be employed to enable asset monitoring. In one such situation, a Cirronet customer desired to reduce overall energy consumption in a large retail and commercial campus. The campus contains a large number of relatively independent stores and venues. The customer believes that providing detailed energy consumption information to the line supervisor of each store, together with benchmarks and targets, would enable the line supervisors to change their behavior and waste less energy. 

The customer made a significant investment to provide the line supervisors the information they would need to manage the energy consumption of their store. Over 150 power submeters were installed across the campus at points consistent with the operations controlled by line supervisors. To reduce the overall project costs, wireless connections were installed between the submeters and the data collection gateways in the campus. Power consumption data is collected from the submeters every fifteen minutes and is stored in a remotely located Aleier EAM. Line supervisors can view their power consumption over a day, a week, a month or any other period they choose. Their data is always presented relative to a benchmark as well as their assigned target for energy consumption for their store. While this project has been implemented too recently to show significant results, the customer expects their annual energy savings to amount to several million dollars. 

Wireless Technologies Are Growing Condition-Based Maintenance

Condition-based maintenance will become more common, and is being increasingly enabled by wireless technology. Condition-based maintenance can help reduce overall maintenance costs while maintaining asset reliability and it is an attractive and feasible strategy for certain types of equipment. Acquiring equipment condition data cost-effectively has been a significant barrier to the adoption of CBM strategies. New wireless technologies are now enabling broader use of condition-based maintenance by lowering the overall costs of these solutions. The benefits of condition-based maintenance are best realized by integrating real-time, continuous equipment data into the overall maintenance management system. These integrated solutions use the native capabilities of the CMMS to identify when maintenance is appropriate, and to manage that maintenance activity as a facet of the overall maintenance operation.

Wayne Stargardt is the Director of Marketing and Product Development at Aleier, Inc. He joined Aleier in 2006 from RF Monolithics and is responsible for marketing, product strategy and product development. Mr. Stargardt has over 25 years of experience in marketing, sales and engineering management in technology companies. Mr. Stargardt was in charge of sales and marketing at SensorLogic, a startup providing M2M (Machine-to-Machine) software as an application service provider. Most recently, Mr. Stargardt was Director of Marketing for RF Monolithics’ Wireless Solutions Group. Mr. Stargardt has a BSME and MSME from the Massachusetts Institute of Technology, a BS from the Sloan School, and an MBA from the Harvard Business School.