The benefits of a data-driven maintenance strategy

Seven steps toward utilizing prognostic data analysis to optimize maintenance schedules.


Seven steps toward utilizing prognostic data analysis to optimize maintenance schedules. Courtesy: CassentecTraditionally, availability forecasts and remaining useful life (RUL) estimations have been based on expert assessments of technical risks and a certain degree of intuition from personal experience with assets. This inside understanding is invaluable. Yet, especially in industries in which downtime is extremely costly new solutions are demanded to assess future risks. Most assets have data readily available - why not using this data to objectively support an expert's experience?

Predictive maintenance solutions offer an excellent starting point to use the data and move towards a data-driven maintenance strategy: based on data analysis it informs in advance that certain equipment will run into trouble soon. Prognostic solutions even go a step further by informing when the equipment will likely run into trouble. Prognostics -- using data analytics to foster expert´s assessment of future availability and Remaining Useful Life (RUL) of assets -- provides help to maintenance engineers in their daily working routine. Operations and asset managers also can make use of Prognostics for a proactive and efficient organization.

The following seven steps are necessary to turn your prognostic data analysis effort into a success:

1. Identification of assets to be monitored
Choose assets that are critical to continuous operation or those that would generate considerable costs or maintenance effort in the case of failure.

2. Identification of key malfunctions
List those malfunctions that reduce efficiency or reliability over time and shorten asset RUL. There is no need for a malfunction history - also malfunctions that have never been observed before for the equipment at hand can be listed.

3. Data review
Visualize the historical condition and process data from the equipment at hand in order to detect trends. Filter out downtime, startup, and shutdown and graph historical condition and process data for each component.

4. Formulation of parameters
Determine what data relates to which malfunctions and the type of the relationship (e.g. correlation or causation). Afterwards, create parameters involving arithmetic or logical functions and dependencies. Finally, check the data patterns for malfunction indicators or explanations.

5. RUL computation
Feed the condition and process data as well as malfunction and parameter specifications into a stochastic model. Project the asset´s condition over an explicit time horizon, apply diagnostic rules at various future time stages, and infer malfunction likelihoods. After obtaining malfunction-specific RUL distributions, total RUL distributions for each asset can be formed.

6. Result validation
Now that the bulk of the work is done, it is time to ask yourself: "Are these results plausible given what we assume or know now? How early can we make a reliable or useful prediction?" This is to check whether the results are plausible and the time frames used are appropriate for conclusive forecasting.

7. Benefit realization
Now you are ready to reap the benefits that are outlined above. You know now how prognostic data analysis works and how you can use it intelligently to improve your maintenance strategy.

Moritz von Plate is the CEO of Cassantec, a company specialized in prognostic Maintenance. Prior to assuming this role, Moritz was CFO and managing director of an EPC contractor specializing in concentrated solar-thermal power plants. Julia Heggemann is the marketing manager of Cassantec. Her responsibilities include marketing, communication and sales. This article is contributed by SMRP, a CFE Media content partner.

Top Plant
The Top Plant program honors outstanding manufacturing facilities in North America.
Product of the Year
The Product of the Year program recognizes products newly released in the manufacturing industries.
System Integrator of the Year
Each year, a panel of Control Engineering and Plant Engineering editors and industry expert judges select the System Integrator of the Year Award winners in three categories.
September 2018
2018 Engineering Leaders under 40, Women in Engineering, Six ways to reduce waste in manufacturing, and Four robot implementation challenges.
GAMS preview, 2018 Mid-Year Report, EAM and Safety
June 2018
2018 Lubrication Guide, Motor and maintenance management, Control system migration
August 2018
SCADA standardization, capital expenditures, data-driven drilling and execution
June 2018
Machine learning, produced water benefits, programming cavity pumps
April 2018
ROVs, rigs, and the real time; wellsite valve manifolds; AI on a chip; analytics use for pipelines
Spring 2018
Burners for heat-treating furnaces, CHP, dryers, gas humidification, and more
August 2018
Choosing an automation controller, Lean manufacturing
September 2018
Effective process analytics; Four reasons why LTE networks are not IIoT ready

Annual Salary Survey

After two years of economic concerns, manufacturing leaders once again have homed in on the single biggest issue facing their operations:

It's the workers—or more specifically, the lack of workers.

The 2017 Plant Engineering Salary Survey looks at not just what plant managers make, but what they think. As they look across their plants today, plant managers say they don’t have the operational depth to take on the new technologies and new challenges of global manufacturing.

Read more: 2017 Salary Survey

The Maintenance and Reliability Coach's blog
Maintenance and reliability tips and best practices from the maintenance and reliability coaches at Allied Reliability Group.
One Voice for Manufacturing
The One Voice for Manufacturing blog reports on federal public policy issues impacting the manufacturing sector. One Voice is a joint effort by the National Tooling and Machining...
The Maintenance and Reliability Professionals Blog
The Society for Maintenance and Reliability Professionals an organization devoted...
Machine Safety
Join this ongoing discussion of machine guarding topics, including solutions assessments, regulatory compliance, gap analysis...
Research Analyst Blog
IMS Research, recently acquired by IHS Inc., is a leading independent supplier of market research and consultancy to the global electronics industry.
Marshall on Maintenance
Maintenance is not optional in manufacturing. It’s a profit center, driving productivity and uptime while reducing overall repair costs.
Lachance on CMMS
The Lachance on CMMS blog is about current maintenance topics. Blogger Paul Lachance is president and chief technology officer for Smartware Group.
Material Handling
This digital report explains how everything from conveyors and robots to automatic picking systems and digital orders have evolved to keep pace with the speed of change in the supply chain.
Electrical Safety Update
This digital report explains how plant engineers need to take greater care when it comes to electrical safety incidents on the plant floor.
IIoT: Machines, Equipment, & Asset Management
Articles in this digital report highlight technologies that enable Industrial Internet of Things, IIoT-related products and strategies.
Randy Steele
Maintenance Manager; California Oils Corp.
Matthew J. Woo, PE, RCDD, LEED AP BD+C
Associate, Electrical Engineering; Wood Harbinger
Randy Oliver
Control Systems Engineer; Robert Bosch Corp.
Data Centers: Impacts of Climate and Cooling Technology
This course focuses on climate analysis, appropriateness of cooling system selection, and combining cooling systems.
Safety First: Arc Flash 101
This course will help identify and reveal electrical hazards and identify the solutions to implementing and maintaining a safe work environment.
Critical Power: Hospital Electrical Systems
This course explains how maintaining power and communication systems through emergency power-generation systems is critical.
Design of Safe and Reliable Hydraulic Systems for Subsea Applications
This eGuide explains how the operation of hydraulic systems for subsea applications requires the user to consider additional aspects because of the unique conditions that apply to the setting
click me