What is predictive maintenance?

Predictive maintenance for industry 4.0 is a method of preventing machine failure by analyzing machine data to identify patterns and predict issues before they happen.  

Until now, factory managers and machine operators carried out scheduled maintenance and regularly repaired machine parts to prevent downtime. In addition to consuming unnecessary resources and costing large sums in forfeited productivity, half of all preventive maintenance activities are ineffective.

Not surprisingly, leveraging industrial IoT technologies for predictive maintenance is a leading Industry 4.0 use case for manufacturers and asset managers. Implementing analytics to monitor asset health, optimize maintenance schedules, and gaining real-time alert to operational risks, allows manufacturers to lower service costs, maximize uptime, and improve production throughput.

 

How does IoT predictive maintenance work?

For predictive maintenance to be carried out, a product or machine requires the following base features:

  1. Sensors – data-collecting sensors installed in the physical product or machine
  2. Communication protocols – the communication system that allows data to flow between the unit and the data store
  3. Data store – the central data hub in which data is stored, processed and analyzed, typically a cloud-based data repository
  4. Data analytics – algorithms applied to the machine data to recognize patterns and generate insights in the form of dashboards and alerts

Machine data is streamed from the sensors to a central repository using communication protocols and gateways. Here, big data analytics techniques are applied to provide valuable insights for reducing downtime and improving product efficiency.
 
How does IoT predictive maintenance work?

Predictive maintenance architecture

 
To implement this system effectively, manufacturers need to map the parameters of failure for machines and create a blueprint for their connected system (the sensors, communication protocols, gateway, cloud, and predictive analytics).

Inside a visual modeling platform, product and engineering teams can graphically define the logic of the system, with a set of logic-based rules that monitor and alert to maintenance issues.

Once in-market, predictive analytics applied to the machine data analyze historic events and compare it back to the blueprint, in order to predict conditions of upcoming failure. A dashboard for predictive analytics synthesizes operational data, allowing manufacturers and operators to visualize the system operation in real-time and derive actionable insights.

 

The benefits of predictive maintenance

Manufacturers and their customers get a range of business benefits from IoT predictive maintenance. The advantages of PdM include:

  1. Reduced maintenance time– Automatic reports for strategic maintenance scheduling and proactive repairs alone reduces maintenance time by 20–50 percent and decreases overall maintenance costs by 5–10 percent. These insights save the manufacturer and their customers time and money.  
  2. Increased efficiencyanalytics-driven insights improve OEE (overall equipment effectiveness) by reducing unnecessary maintenance, extend asset life and enable root cause analysis of a system to uncover issues ahead of failure.
  3. New revenue streams- Manufacturers can monetize industrial predictive maintenance by offering analytics-driven services for their customers, including PdM dashboards, optimized maintenance schedules, or a technician dispatch service before parts need replacement. The ability to provide digital services to customers based on data presents an opportunity for recurring revenue streams and a new growth engine for companies.
  4. Improved customer satisfaction- Send customers automated alerts when parts need to be replaced and suggest timely maintenance services to boost satisfaction and provide a greater measure of predictability.
  5. Competitive advantage- Predictive maintenance strengthens company branding and value to customers, differentiating their products from the competition and allowing them to provide continuous benefit in-market.

 

Take the free whitepaper with youWhy predictive maintenance is driving industry 4.0

 

Predictive maintenance tools

Once a manufacturer has chosen the right sensors, cloud, communication protocols and analytics platform for their products or machines, implementing predictive maintenance requires a baseline of common tools.

Predictive maintenance tools include an IoT development platform to model, test and deploy the predictive maintenance solution, data analytics algorithms to detect patterns based on historical and current machine data, and workflows to feed the data into existing ERP and CRM systems.

 

What is the difference between preventive and predictive maintenance?

Manufacturers have been carrying out different forms of preventive and predictive maintenance for years. Understanding the difference between them, however, is critical with the emergence of Industry 4.0.

Preventive maintenance depends on visual inspections, followed by routine asset monitoring that provide limited, objective information about the condition of the machine or system. In this process, manufacturers regularly maintain and repair a machine to prevent failure.

On the other hand, PdM is data-driven and relies on analytics insights for strategic maintenance and repairs ahead of disruptions in production.

 

How are companies using IoT predictive maintenance tools?

Organizations are implementing predictive maintenance analytics in a range of ways, from targeted solutions for a single machine part, to factory-wide deployments for increasing OEE throughout the production line.

For machine and parts manufacturers, a relatively common predictive maintenance use case is monitoring and analyzing the condition of a motor to get alerts about its productivity levels, power consumption, health status, and internal wear.

Manufacturers are also turning to predictive maintenance for Factory 4.0, or a connected factory, by installing sensors in machines, workstations, and other designated sites such as the HVAC, security cameras or worker equipment, to predict issues across the factory floor.
 

predictive maintenance use cases

 

Common approaches to IoT predictive maintenance

The two most common approaches to predictive maintenance are rule-based and machine learning-based.

Rule-based predictive maintenance

Also referred to as condition monitoring, rule-based predictive maintenance relies on sensors to continuously collect data about assets, and sends alerts according to predefined rules, including when a specified threshold has been reached.

With rule-based analytics, product teams work alongside engineering and customer service departments to establish causes or contributing factors to their machines failing.

Once common reasons for product or part failures are established, manufacturers can build a virtual model of their connected system. Here they define product use cases, with “if-this-then-that” rules which describe the behaviors and inter-dependencies between the various IoT system components.

For example, if temperature and rotation speed are above certain predefined levels, the system will send an alert to an operator dashboard, to address the issue ahead of failure.

These rules provide a level of automated, predictive maintenance, but they are still dependent on a product team’s understanding of what parts or environmental elements require measuring.

The condition monitoring dashboards can be integrated with insight from machine learning to provide a visually understandable heatmap of asset conditions in real-time.

Predictive Maintenance Machine Learning

Predictive maintenance with machine learning looks at large sets of historical or test data, combined with tailored machine-learning algorithms, to run different scenarios and predict what will go wrong, and when.

Predictive Maintenance Algorithms

In this scenario, algorithms learn a machine’s normal data behavior and uses this to identify and alert to deviations in real-time.

The algorithms required for machine learning must analyze input (historical or a training set of data) and output data (the desired result). A machine monitoring system includes input on a range of factors from temperature to pressure and engine speed. The output is the variable in question – a warning of a future system or part failure. The system will then be able to predict when a breakdown is likely to occur.

 

How can I offer new customer service with predictive maintenance?

Predictive maintenance offers OEMs with a business value proposition to offer data-driven customer services and get recurring revenue in return. With PdM services, companies can build a subscription-model for customers to access dashboards or reports that will improve their OEE and reduce maintenance costs. Additional opportunities for services can be found in dispatching technicians for repairs, setting up alert systems, and shipping parts that need replacement ahead of failure.  

 

How do I get started with IoT predictive maintenance?

Predictive maintenance is often restricted to the minority of companies whose machines have been collecting data for years, and can utilize advanced analytics platforms to sort through the data with machine learning.

But for companies implementing a connected system for the first time, with the ultimate goal of implementing machine learning for predictive maintenance, a pragmatic way to get started with IoT predictive maintenance.

With rule-based PdM, manufacturers can bypass the need for a large historical data set or advanced machine learning algorithms at the outset.

 

Predictive Maintenance Machine Learning

 

This gives companies quick business results and a stepping stone into machine learning. In this model, product teams start with basic assumptions or ‘rules’ based on ‘what if’ scenarios that can be easily defined, rather than a machine algorithm running possible scenarios.

Rule-based PdM is achievable, affordable, and delivers measurable business benefits. The easiest way to get started is with an IoT platform centered on a rule-based model, which enables teams to quickly define, simulate and deploy a predictive maintenance solution for their products.