What is predictive maintenance?

Predictive maintenance technologies are designed to assist in determining the state of in-service equipment so that maintenance can be scheduled. Predictive maintenance is the application of information; proactive maintenance approaches examine the condition of equipment and anticipate when it should maintain. The purpose of predictive maintenance is to forecast when equipment will fail (depending on a variety of parameters), then prevent the failure through routine and corrective maintenance.
Condition monitoring is the continual monitoring of machines during process conditions to maintain optimal machine use, which is necessary for predictive maintenance. There are three types of condition monitoring: online, periodic, and remote. Finally, remote condition monitoring allows the equipment observed from a small place and data supplied for analysis.

Importance of predictive maintenance

The figure shows the importance of predictive maintenance, the figure consist of reactive, planned, proactive, and predictive maintenance.
Reactive maintenance is required when equipment breaks. In planned maintenance, there is a maintenance schedule provided for equipment. Proactive maintenance is detecting faults to improve the efficiency and performance of the equipment. And predictive maintenance consists of advanced analysis and sensing data to increase machine reliability.

Condition monitoring is required for predictive maintenance to work. Ensure asset optimization and machines conduct constant monitoring in real-world situations. Predictive maintenance, like any other maintenance plan, aims to:

• Improve asset dependability to reduce breakdowns and increase asset uptime.
• Reduce maintenance effort to reduce operational costs.
• Reduce maintenance expenditures and increase production time to improve maintenance budgets.

Technology of predictive maintenance/predictive maintenance strategy

There is not a single system that can handle all aspects of predictive maintenance. On the other hand, manufacturers employ a variety of condition-monitoring equipment and procedures to efficiently predict failures and raise red lights when maintenance is required.

1. Infrared thermography: Infrared thermography is a non-invasive testing technique commonly employed in predictive maintenance. Maintenance employees can detect above-normal temperatures in equipment using infrared cameras. Components worn out or have faulty circuits heat up, which shows up as a heat spot on a thermal image. Infrared inspections can detect these hotspots early on and fix equipment, minimizing the likelihood of greater problems developing. Infrared technology is a versatile tool that may apply to various machinery and infrastructure projects.
2. Acoustic monitoring: Acoustic monitoring at the sonic and ultrasonic ranges can help maintenance personnel detect gas, liquid, or vacuum leaks in equipment. Ultrasonic technology has more applications than sonic technology and is more expensive, but it is far more reliable for machines. These technologies, of course, are a complement to the technician's most valuable tool: their ears. Regular listening can be used in conjunction with sonic and ultrasonic technology to determine why a gearbox sounds odd or if there is a possible leak.
3. Vibration analysis: High-speed rotating machinery is subjected to vibration analysis. A technician employs handheld devices or real-time sensors built into the equipment to monitor equipment performance. When a machine functions at its best, it produces a distinct vibratory pattern. The vibration alters as the components wear down, and a new way arises. With continuous monitoring, a competent technician can match vibration pattern readings against known failure scenarios and resolve an issue sooner. Vibration analysis can discover misalignment, out-of-shape shafts, unbalanced elements, loose mechanical components, and motor problems. Because vibration analysis is difficult to predict, technicians must be well-trained. The high expense of vibration analysis is the main roadblock.
4. Oil analysis: In predictive maintenance, oil analysis is a useful tool. Technicians can detect impurities in oil by inspecting the state of the oil. The acidic or basic number is determined by determining oil's viscosity, water, and particle counts. The primary advantage of oil analysis is that the results of the initial tests serve as a benchmark for any new machinery or maintenance.

Working of predictive maintenance

Before implementing a predictive maintenance program, an organization must take several steps, including:

• Examining the requirements as well as the existing equipment.
• We look at all downtime, equipment failures, losses (yield and energy), regulatory fines, and workplace safety data.
• Defining terms and concepts, as well as presenting a case for Predictive Maintenance(PdM).
• Getting buy-in from key stakeholders and educating them.
• Completing an inventory of equipment and assessing its current condition.
• Choosing equipment for the first phase of the program's implementation.
• System details are being developed depending on particular systems and/or components.
• Taking a look at any preventative or predictive maintenance that is already in place.
• Choosing which systems to include and what to look for during the inspection.
• Defining the criticality of the program and determining the PdM frequency and schedule type.
• assessing predicted resources and assigning roles and tasks to persons.
• Organizing and integrating the program into the scheduling system.
• Educating and gaining support from the operations and maintenance departments.
• Purchasing new equipment and providing training.
• Creating a CMMS (computerized maintenance management system).

Advantages of predictive maintenance

• Maintenance costs are reduced: Predictive maintenance can help reduce maintenance costs. That is especially critical when businesses must invest in personnel, maintenance, replacement parts, tools, and equipment in the event of a significant failure.
• Machine failures are reduced: Machine failures have been the subject of a lot of research. Machine and system monitoring regularly can help to reduce the likelihood of large-scale failures. The frequency and nature of machine breakdowns usually decrease after two years of establishing a predictive maintenance program.
• Reduced Downtime: Repairing equipment requires less time with predictive maintenance. Regular monitoring and evaluation of machine conditions in swiftly locating and repairing defective components on all equipment is done. This cuts downtime and, in many cases, eliminates it.
• Increased Machine Lifespan: It is possible to extend the life of machinery by detecting problems early (before they become catastrophic failures). A condition-based predictive maintenance approach ensures that equipment is never damaged to the point of failure. The longer the equipment lasts, the better the return on investment for the company.
• Stocking Reduction: Companies frequently have to cope with big stock investments in multiple sectors, which might cause capital to be locked up. If the parts are not used as soon as possible, their quality will decline, and they may be thrown away. Rather than keeping a big stock of components on hand, ordering parts just when needed can save money on stocking.
• Increased Production: Condition-based predictive maintenance solutions must be backed up by reliable process systems to be effective. A complete predictive program that includes parameter monitoring can boost operational efficiency and, as a result, productivity.
• Enhanced Operator Security: Early warning signs can be installed with predictive maintenance to prevent injuries caused by defective machinery. Manufacturers who adopt a condition-based predictive maintenance program are often recognized and offered advantages by insurance companies. This program can help you save money on insurance without sacrificing coverage.

Context and Applications

• Drones and sensors that map utility networks help prevent energy disruptions.
• Detect a temperature drop in a steam pipeline, indicating a pressure leak.
• Increased temperatures in electrical panels should capture to avoid component failure.
• For power consumption monitoring, measure supply-side and demand-side power at the same connection point.
• Overloads in electrical panels should be located.
• Identify spikes in motor amperage or overheating caused by faulty bearings or insulation breakdowns.
• Find three-phase power imbalances caused by harmonic distortion, overloads, phase degradation, or failure.

The topic is related to the following courses:

• Bachelor of Technology in Electrical Engineering.
• Master of Technology in Electrical Engineering.

Related Concepts

• Preventive Maintenance
• Predictive Maintenance Algorithm
• Predictive Maintenance Technique
• Predictive Maintenance - Machine and data

Common Mistakes

1. Poor data and planning: Before starting the predictive maintenance process, should have all the reports that are maintenance records, breakdown data, also need to check original manufacturing suggestions, which is a useful guideline to schedule predictive maintenance.
2. Selecting the incorrect equipment to perform Predictive maintenance: Before starting the predictive maintenance schedule, the first step should be to choose the correct equipment, and when this is done, that will save the money.
3. Maintenance workers are poorly trained: Most of the time, many projects are failed due to improperly trained staff or workers. If maintenance workers are not well prepared, then be told about a new concept and best predictive maintenance methods.
4. Poor software quality is used: If the maintenance strategy/technology is improperly designed and difficult to use, it can result in performance problems, member of the team de-motivation, and change resistance.
5. Attempting to make everything predictable.
6. Data is being collected, but it is not being analysed.
7. Immediate Results are expected.
8. Data that has been collected has not been integrated.
9. Technicians were not properly trained to use the tools.

Practice Problems

Q1. Condition monitoring is divided into .........

A. Online

B. Periodic

C. Remote.

D. All of the above

Answer: D

Explanation: Condition monitoring is divided into three parts that is Online, Periodic and Remote.

Q2. "The continuous monitoring of machines or manufacturing processes, with data collected on critical speeds and changing spindle locations." is called as......

A. online conditioning monitoring

B. periodic conditioning monitoring

C. remote conditioning monitoring

D. None of the above

Answer: A

Explanation: It is the definition of online conditioning monitoring.

Q3. ..................... "gives insight into changing vibration behavior of installations."

A. online conditioning monitoring

B. periodic conditioning monitoring

C. remote conditioning monitoring

D. None of the above

Answer: B

Explanation: Periodic condition monitoring, which is accomplished by vibration analysis.

Q4. Technology of predictive maintenance........

A. Infrared Thermography

B. Acoustic Monitoring

C. Vibration Analysis

D. Oil Analysis

E. All of the above

Answer: E

Explanation: These are techniques used in predictive maintenance.

Q5. High-speed rotating machinery is subjected to..............

A. Infrared thermography

B. Acoustic monitoring

C. Vibration Analysis

D. Oil Analysis

Answer: C

Explanation: Misalignment, out-of-shape shafts, unbalanced elements, loose mechanical components, and motor difficulties can all be detected via vibration analysis.