Understanding Plant Asset Management Systems

Comprehensive design links control systems, predictive maintenance systems, CMMS, and enterprise systems to support equipment asset management decisions.

Competition in the global economy has put industrial enterprises under intensive pressure. There have been significant shifts in the relationships between producers, suppliers, and consumers. The need for improved production reliability and reduced expenses is clearly demonstrated by production strategies that include just-in-time material supply and delivery. Both suppliers and consumers are working to optimize their cash flow by managing throughput and reducing the expense associated with maintaining excess inventories, while still ensuring that their product throughput requirements are met.

From just-in-time delivery to higher quality and increased technical support, customers are requiring more from their suppliers. Not only must the product be of high quality, and at the lowest possible price, but deliveries must be on time. Often severe financial penalties are imposed by an industrial partner consumer when a supplier fails to deliver on-time or at required quality thresholds. Consequently, the financial impact of unexpectedly stopping a production line or discarding a batch of a product can be devastating.

Because of the need to ensure that production commitments are achieved, companies increasingly are turning to plant asset management as an optimization strategy to improve their process efficiency and reduce maintenance, thus enhancing their return on assets. According to a June 1999 study by the ARC Advisory Group, companies are reporting as much as a 30 percent reduction in maintenance budgets and up to a 20 percent reduction in production downtime as a result of implementing a plant asset management strategy. Since as much as 40 percent of manufacturing revenues are budgeted for maintenance, these savings contribute significantly to a company's bottom line.

Maintenance strategies that once were "run-to-failure" now are "condition-based." Enterprise asset management (EAM) systems and computerized maintenance management systems (CMMS) are implemented to support maintenance scheduling, workflow management, inventory management, and purchasing, and to integrate these functions with automation, production scheduling, and manufacturing systems. Leading corporations now have direct connections from their EAM system to electronic-commerce maintenance, repair, & overhaul/operations (MRO) procurement systems, which allow paperless purchasing of parts and offer considerable time and cost savings compared with traditional purchasing methods.

0700mm21
Plant asset management (PAM) systems tie into a variety of plant information systems.
Purpose of a plant asset management system
Since a critical factor in both maintenance and operational scheduling is the ability to constantly monitor the health of plant assets, corporations now are implementing complete plant asset management (PAM) systems. A PAM system allows plant personnel to assess the risk of premature production outages and the ability to schedule and plan future maintenance activities. Let's explore the purpose of a PAM system in a little more detail and then discuss various components of a PAM system.

The purpose of a PAM system is to provide timely information to operations and maintenance (O&M) personnel in order to safely increase the total production output of a plant at a reduced cost per unit of output. These benefits occur as the manufacturing facility makes optimum operating and maintenance decisions through the application of a PAM system's information solution. O&M personnel are constantly faced with decision-making based on limited information. PAM systems make this decision-making job easier by providing knowledge about the current and future condition of vital production assets.

Maintenance support
PAM systems assist maintenance personnel in answering the following questions:

  • What equipment may fail if it does not receive maintenance intervention?
  • What intervention should we take and how soon?
  • What parts should I order and how soon?
  • What is the optimal blend of condition-based (CBM), calendar-based (PM), usage-based (PM), and run-to-failure maintenance for a given piece of equipment?

Operations support
PAM systems assist operators and production planning personnel in answering the questions:

  • Should I make any adjustments to my process now to prolong the life of assets critical to my process?
  • To what extent can I increase my process output without incurring an unacceptably high risk of unexpected process slowtime, downtime, quality problems, or safety shutdowns?
  • What is the risk of successfully producing X amount of product next week given a projected process utilization rate of Y?

The role of the PAM system as it turns plant measurement data into actionable information and issues advisories to both maintenance and operation systems by synthesizing the asset measurements it has obtained is outlined in the accompanying diagram "Plant Information Systems." A PAM system contains eight modules: asset information register, data harvester, computed indicators calculator, data archiver, condition monitor, asset health analyzer, O&M advisory manager, and O&M gateway manager.

Asset information register
The first module of a PAM system is the asset information register. This module provides the rest of the PAM modules with information about the location of the asset and its criticality to the process, as well as asset-specific model data and nameplate information. Registers also need to store measurement location information, such as the type of transducer being used, the post-processing to perform on a measurement location, and the spatial orientation of orientation-sensitive measurements such as vibration locations. Some registers also keep information from a reliability study such as an RCM audit, as well as financial metrics that could influence decisions regarding the asset. Others include the dates of future maintenance tasks, such as planned overhauls, and can track work and failure histories on the asset through gateways to external systems.

Companies interested in open systems should look for register modules which support open asset information standards from open industry alliances such as the Machinery Information Management Open Systems Alliance (MIMOSA) which now publishes a universal set of codes for all asset equipment types, asset nameplate data, and measurement location types.

Data harvester
The next module of a PAM system is the data harvester. This module periodically gathers data from off-line and on-line measurements on assets ranging from smart valves to large turbines. In the off-line area, the harvester module contains interfaces to load and unload route-based schedules to various walk-around data collection devices, such as vibration data collectors, as well as operator inspection log devices and manually-entered inspection data related to assets. In the on-line area, the harvester module periodically extracts data from turbo-machinery protection monitoring systems, high-speed transient monitoring systems, periodic surveillance monitoring systems, control device monitoring systems, and process data historians.

The harvester synthesizes data from various monitoring technology systems, including shaft displacement, casing vibration, ultrasonic, electrical circuit, thermographic imaging, oil particulate, and oil chemical analysis systems. It correlates this condition-based monitoring data with the current process data in order for the PAM system to properly associate the dependent variables, such as vibration, with the independent variables, such as speed and load.

For companies looking for open systems, some suppliers of PAM data harvester modules now support open plant data access standards such as MIMOSA's Tech-File Import and Tech-XML Client interfaces (supporting both dynamic and scalar current value and historical data) and OPC Foundation's Data Access Client interface (for current scalar current value data only). These interfaces allow a harvester module to access any monitoring or measurement system that supports these universal data access standards.

Computed indicators calculator
A PAM system's computed indicators calculator module derives "features" to be extracted from the raw measurements and dynamic spectra as well as calculates "macro" indicators derived from multiple measurements (such as differential pressure). Calculations of rotating shaft and bearing vibration, sound, and electrical frequencies allow for a sophisticated "fingerprint" analysis of dynamic frequency data. These computed indicators are vital to properly discover early abnormalities in an asset. Some computed indicators calculator modules allow for the definition of a virtual sensor measurement, which is trended and treated as a physical measurement reading.

Data archiver
A PAM system's data archiver module provides long-term data storage of plant measurements with options for data error flagging, compression, and expiration. Data error flagging techniques include the tagging of data "outlyers" as well as the tagging of data with known data collection errors or data taken when the asset was off-line. Sophisticated data archivers also manage compression of the data by defining a "dead-band" range where a new data point must cross in order to be permanently stored. Archivers also manage data expiration and allow physical deletion from the on-line database, though many plants prefer to keep data in the archiver for up to 5 years in order to look at long-term trends in asset condition monitoring and performance data.

State-of-the-art archivers utilize industry-standard relational databases, such as Oracle and Microsofts SQL Server, and allow external access for distributed database management and other database administration functionality.

In the open systems arena, MIMOSA publishes an open Tech-XML Server and Tech-SQL Server interface (supporting both dynamic and scalar historical data) for data achivers (dynamic and scalar historical data) and the OPC Foundation publishes an OPC Historical Data Access Server interface (scalar historical data only) to allow archivers to "serve up" their data to other systems that support the same universal data access standards.

Condition monitor
A PAM system's condition monitor module facilitates the creation and maintenance of an asset baseline "profile" and then searches for abnormalities whenever new data or indicators enter the PAM system. The condition monitor module allows the end-user to establish normal and abnormal conditions for all measurements and computed indicators in the database. The measurements of interest can range from simple parameters such as temperature and oil particle count to complex data such as vibration spectra or infrared images. In all of these cases, the objective is to determine what is normal for the machine and identify the equipment in various abnormal "alarm" states.

Advanced PAM systems include inputs from process control data historians and sophisticated state-aware condition monitoring technology that can automatically set multiple "baselines" for equipment based on variable operating loads, speeds, and other process conditions. This allows the system to be sensitive to the current operational "state" so as not to over-alarm or under-alarm.

Asset health analyzer
The asset health analyzer acts on exceptions found by the condition-monitoring module. The analyzer module facilitates and permanently archives an analyst's evaluation of the current health of the asset in question. This process is assisted by integrating all relevant data into information displays that allow multi-disciplinary data (lubrication, vibration, thermographic, ultrasonic, process data, etc.) to be visually compared in multi-parameter plots and graphs. This asset health analysis is aided also through the use of automated diagnostic tools and rule sets.

After performing a diagnosis, a prognostic assessment is also needed to determine the future health of the asset in question and its projected time to failure and failure mode. An additional prognostic assessment also is required if the asset's failure mode will cause an impact on operations. This step is aided by tools that provide for easy review of the asset failure database, criticality analysis, failure modes and effects analysis, risk-based monitoring data, reliability-centered maintenance studies, and other reliability data. If production will be affected, then the asset health analyzer needs to store the projected process time to failure and failure mode.

To summarize, the analyzer records and stores the following output from the human analyst or diagnostic system:

  • What data is truly abnormal for the process conditions? (asset symptoms)
  • What could be causing the abnormality? (asset diagnosis)
  • How and when will the asset fail if no action is taken? (asset prognosis)
  • How and when will the process fail if no action is taken? (process prognosis)

In the area of open standards for asset health analyzer modules, MIMOSA publishes a universal set of codes of symptoms, diagnoses, and prognoses which assists companies who purchase MIMOSA-based PAM analysis systems to "mine" the asset health analyzer database to better understand which problems are being properly diagnosed and which ones are being overlooked.

0700mm22
A complete PAM system contains eight modules.
O&M advisory manager
After all of an asset's health problems have been diagnosed and their future impact assessed, the PAM operations and maintenance (O&M) advisory manager facilitates the creation and permanent storage of operations and maintenance advisories generated by a human or automated expert. To aid an analyst, the advisory manager can easily retrieve advisories previously issued for the same diagnosis from a similar class of equipment. A priority code should be assigned to the advisory and the source of the advisory stored.

O&M gateway manager
The final module in a PAM system completes the transformation of data into actionable information. The O&M gateway manager module creates and manages gateways between the PAM system and a plant's operations and maintenance systems and personnel.

Most plants now have an EAM system or computerized maintenance management system (CMMS) which manages maintenance work management and parts management. The PAM gateway manager module establishes connectivity with the EAM/CMMS system. It issues and tracks work requests based on the asset's health analysis and retrieves work history information to facilitate better diagnosis and advisories. State-of-the-art gateway manager modules can submit work requests directly into an EAM/CMMS system, monitor the progress of the work order, and view equipment work histories in a table format or in a graphical Gantt chart. This allows the analyst to make better diagnostic decisions and to see the impact of maintenance on the condition of a piece of equipment. The EAM/CMMS then should be able to act on this information to order spare parts or issue a work order for repair, overhaul, or further monitoring.

PAM O&M gateway manager modules also require connectivity with the plant's automation systems in order to issue operational alarms and operation change requests. The modules should communicate urgent asset health alarms and recommendations to the human-machine interface (HMI) displays from a plant automation system (PAS) or distributed control system (DCS), which the operators are constantly reviewing. In addition, the module also should send operational change requests to the planning module of a manufacturing execution system (MES) in order to affect operational changes to extend a critical asset's useful life, thus optimizing production throughput and quality.

In order to communicate directly with O&M personnel, most state-of-the-art O&M gateway managers include e-mail and paging interfaces in order to notify plant personnel of urgent, impending asset failures. E-mail or paging message templates can be designed and then utilized when a given asset health condition warrants a transmission.

New XML-based integration standards from MIMOSA (Work-XML and Reg-XML Server) will allow bi-directional gateways to be built using universally, open systems. The XML framework allows for interoperability across intranets or even on the Internet. Companies desiring open systems should consider utilizing these industry standard interfaces wherever feasible.

Case study
Baltimore Gas & Electric Co. (BGE) is the first U.S. gas utility and one of the earliest electric utilities. In 1991, anticipating deregulation of the utility industry, BGE began developing a foundation to perform as a profitable, world-class energy company. To compete effectively in the deregulated market, BGE knew it needed to minimize costs and at the same time maximize the plants' power generation capacity. Since maintenance is one of a utility's largest controllable costs, the company began by looking for ways to improve operational efficiency and reduce operations and maintenance costs.

BGE implemented a plant asset management system in 1993 that integrated condition information from across all eight fossil-fuel power plants. BGE embraced a wide range of test technologies (oil analysis, vibration analysis, motor monitoring, etc.) in its facilities. The company documented cost savings of $16 million between 1994 and 1998 from this PAM system technology utilizing predictive maintenance to optimize operations and maintenance. In addition, it increased production by 14 percent while reducing fuel costs and improving heat rates.

Manufacturing and production enterprises are under intense pressure to achieve maximum efficiency. The winners will be those that maximize their investment in people and equipment assets to achieve highest profitability. For physical assets, the objective is to optimize the utilization of all plant assets—from entire process lines to individual pressure vessels, piping, process machinery, and vital machine components. The use of PAM systems is now making this a reality for state-of-the-art plants today. MT


This e-mail address is being protected from spambots. You need JavaScript enabled to view it is strategic project manager for Entek, 1700 Edison Dr., Milford, OH 45150; (513) 576-6151; Internet www.entek.com, and technical director for the Machinery Information Management Open Systems Alliance .