The Benefits of a Predictive Maintenance Plan

In the age of Industry 4.0 and the Industrial Internet of Things, plant maintenance has the potential to become less expensive and disruptive, and more insightful. To achieve these benefits, however, may require a change in how you maintain your plant’s equipment. 

Predictive Maintenance

There are four main ways you may currently be managing your plant’s equipment:

  1. Reactive maintenance – replacing parts only when machinery or equipment breaks.
  2. Proactive maintenance – identifying and addressing problems that lead to breakdowns.
  3. Preventive maintenance – maintaining equipment in good operating order to prevent breakdowns.
  4. Predictive maintenance – using inputs from connected machines and devices, remote monitoring, predictive analytics, and automated maintenance orders to predict when maintenance is necessary based. Predictive maintenance can also be done manually by periodically checking equipment with handheld vibration or audio sensors.

To help you determine how predictive maintenance can offer greater benefits and transform your plant’s operational costs, let’s start with an example that lets you weigh the costs and benefits of each maintenance approach. Ultimately, as you’ll see, it will be a combination of proactive, preventative, and predictive maintenance that will help you achieve the best results.

The Costs and Benefits of Reactive, Proactive, and Preventative Maintenance

Assume that somewhere in your plant right now, a bearing is going bad. How will your maintenance organization deal with it?

If your plant is like many, the first sign of an impending failure is a change in how the machine sounds, which an operator will often notice. Or assuming it’s a bushing that’s failing, the operator might see metal shavings. But if these signs go unnoticed, you may face catastrophe, as the part vibrates, heats up, or breaks. The typical response is reactive maintenance—where you call in a technician to restore equipment to its normal operating condition. Although costs are low until machine failure, the end result is that you incur a much greater expense for unexpected downtime. All the while, your backlog of orders is piling up. And, you may wait a long time for spare parts, unless you’ve stockpiled just the right ones.

With sufficient data, you may also be able to employ proactive maintenance to identify and address the problems that lead to breakdowns. For the bearing, the issues could be inappropriate lubrication, poorly aligned parts, or unfavorable environmental conditions. You’ll need to equip your plant with sensors or employ handheld products to locate areas where you might place sensors to supply data for a predictive or preventive plan. If you have the data to address root causes of parts failures, proactive maintenance can help prevent equipment failures and resulting downtime, disruption, and cost.

To move past the antiquated method of running equipment to failure, you may try to maximize uptime with preventive maintenance. You schedule maintenance based on time estimates for components from historical records or OEM recommendations. Although preventive maintenance is a step up from reactive maintenance, it also has some drawbacks. You may end up servicing and replacing good parts, while increasing downtime and disrupting operations to check equipment that’s running just fine. That’s where predictive maintenance can help.

The Value of Predictive Maintenance

With predictive maintenance, sensors can gather relevant, real-time data on the condition of the bearing. This information is then stored in a secure, cloud-based network that you can access at any time. Data from the sensor can be used to predict when you should perform proactive maintenance or preventative maintenance.

To minimize disruption, most predictive maintenance can be performed while equipment is operating. Machine-to-human communication and data analytics give you insight into performance levels without manual routine check-ups.

In recent years, costs have decreased on technologies such as sensors, computing power, data storage, and bandwidth, allowing even smaller organizations to get started with predictive maintenance. Still, outfitting your plant requires an initial investment that can be intimidating. But, there’s good news. Although entry level costs may be high, you can expect to see a 10X return on investment in predictive maintenance within the first two years and a reduction in maintenance costs of 25 to 30 percent according to the U.S. Department of Energy. A study by the Enterprise Strategy Group found that unplanned downtime for the average manufacturing plant can cost $30,000 to $50,000 per hour.

Sensors that have a say!

JUMO IO-Link sensors for temperature and pressure measurement

Long plant downtimes now belong to the past. The new JUMO temperature and pressure sensors with IO-Link use the integrated diagnostic function to better plan the availability or the exchange of sensors. In addition, time-consuming parameterizations when changing sensors are eliminated as the necessary data is transferred from the superordinate system.

Your benefits in a nutshell

  • Optimization of the production process through communication down to the lowest field level
  • Reduction of mounting and startup times
  • Increase of plant efficiency due to maximum transparency down to the sensor level
  • Reduction of maintenance and service costs with simultaneous increase in plant availability
  • High degree of process reliability due to long operating life and great accuracy
  • Flexible use through compact design type and a variety of process connections

What is IO-Link?

IO-Link is simply flexible – optimization of the production process through communication down to the lowest field level

Flexibility, production process optimization, and remote serviceability are important performance parameters for machines or plants. Sensors with IO-Link now give the plant operator access to the lowest field level. Only minimal effort is required to retrieve sensor information, parameterization, and diagnoses so that plant conditions can be ideally evaluated. The efficient point-to-point communication of IO-Link is based on the well-known three-wire sensor connection that does not place additional demands on the cable material. IO-Link is consequently not a fieldbus, but the further development of the previous and proven connection technology for sensors. IO-Link is a serial, bidirectional point-to-point connection for signal transmission and energy supply within any number of networks, fieldbuses, or back panel buses.

IO-LINK is quick and straightforward – reduction of mounting and startup times

The use of sensors with IO-Link can significantly reduce the required effort for mounting and startup. This advantage is made possible through simplified cabling as well as automation of the startup through parameter storage and duplication. During mounting, ready-to-install cables are used so that no assembly is required and error sources are omitted. Startup is also possible through automation as the parameters can be downloaded and made available to the device in a matter of seconds. Users particularly appreciate IO-Link due to its simple installation and parameterization as well as its independence from the fieldbus. The result is that the need for wiring is significantly decreased and that each sensor always has its own “ID card” due to consistent parameter data retention. This greatly reduces the amount of work involved in troubleshooting.

IO-Link is simply efficient – increase of plant efficiency due to maximum transparency down to the sensor level

Times of unexpected plant failures due to a sensor defect are over. The reason here is that the integrated diagnostic mechanisms allow early recognition and repair of defective sensor states. The functions contained in the sensors – such as operating hours counter, drag indicator, and detection of probe breaks/short-circuits – help to evaluate the sensor states early enough to react so that plant efficiency increases considerably. IO-Link offers the option of exchanging cyclical as well as acyclic data with superordinate levels. For example, parameter data can be downloaded to a sensor or, alternatively, diagnostic data can be extracted during operation. Due to a transmission speed COM 3 with 230.4 kBaud and the cycle time of 2 ms data is quickly exchanged and available within seconds.

IO-Link has an eye on costs – cost reduction while plant availability increases at the same time

IO-Link closes the communication gap between the field level and the sensor level. As a result, other than the process variables, data for identification, for parameterization, and about the condition of the device can be transferred. Now information is available that prevents the mix-up with wrong device types during device exchange. The parameterization of the sensors is saved in a superimposed fashion so that it can be automatically transferred during device exchange.
Each IO-Link devices includes a device ID. The IO-Link master retrieves the device ID and can assign the device to an IODD. This offers the option to distinguish the sensor type (temperature sensor/pressure sensor) from other ones through the device id as each sensor type possesses several device IDs. These clearly identify the sensor and describe its different features. One result is that the exchange of a sensor with another that deviates in the measuring range and accuracy class can be identified immediately. The wrongly mounted sensor is instantly replaced by the correct one and is not even implemented, which prevents its destruction or an error during running operation of the plant.