The Basis of Industrial IoT- Condition Monitoring

Condition Monitoring (CM) Defined

The term Condition Monitoring is self-explanatory and it basically means to monitor an asset’s condition. CM is seen as the foundation of what is known as Industry 4.0.

An integral function of CM within the IIoT ecosystem is to provide data that can ultimately be used for various smart factory applications, including Digital Twins and Predictive Maintenance (PdM).

Condition Assessment vs Condition Monitoring

A momentary snapshot of a component’s health is taken when a technician does a routine visual check of any component in a plant. This is normally pre-scheduled and does not take into consideration previous inspections or historical performance data. This kind of inspection is called a Condition Assessment.

Condition Monitoring takes a much wider set of granular data into account. This includes historical trends, condition and location of the plant, other components of the same type, previous inspections and sensor data from the asset. The analysis done not only determines the component’s present status but also forecasts future problems and when they are likely to happen, including when the part needs to be replaced.

5 Benefits of IoT Condition Monitoring

Condition Monitoring has several benefits for a business, including the advantages of reducing resources and costs.

The major benefits of condition monitoring are:

1.     Maximizing production

New productivity levels can be reached by using extensive and accurate readings from production machine sensors, combined with visibility into production inefficiencies by using data analytics algorithms. This is particularly true where condition monitoring is used in the oil and gas industry.

2.     Reduction of maintenance costs

Repairs are performed before critical damage occurs and labor and travel costs are cut as maintenance is done proactively and timely. A reduction in service time, and improved customer satisfaction are also direct benefits of Condition Monitoring.

3.     Product development is driven by relevant and accurate data

Data collected on asset behavior over time can be analyzed by engineers and help them identify design flaws in a product that can then be corrected in later product versions.

4.     Optimizing spare part inventory

Rather than running out of spares, thereby increasing downtime, or overstocking on expensive spare parts and impacting margins, Condition Monitoring allows for accurate forecasting of spare part demand.

5.     Extending the lifetime of machinery

Detailed monitored of the health of machines and all of their components is done. Things like wear-and-tear, overheating, and other threats to a machine’s health are resolved in a timely fashion, thereby lengthening the lifespan of a machine.

Techniques used in Condition Monitoring

Different manufacturer implement condition monitoring differently. This is due to every asset or product having its own unique normal behavioral pattern which should be monitored and analyzed.

Some commonly used Condition Monitoring techniques include:

  • Lubricant analysis
  • Vibration analysis
  • Acoustic emission
  • Infrared thermography
  • Motor current signature analytics (MCSA)
  • Ultrasound
  • Model-based voltages and currents (MBVI)

One example would be running a specified voltage through a motor, after which the current is measured. This data is then compared to that of a mathematical model that has as an input the real-time accurate data from the same motor. The two current values are compared after having been summed. If there are no deviations present, the motor would be seen as healthy. If there are however discrepancies, an analysis is done to determine what the problem is. Once a problem has been identified, it can be classified and the relevant solution deployed.

This example clarifies the durability of Condition Monitoring. It is logical to constantly monitor and record the status of the motor, rather than periodically perform a diagnostics check. This allows for historical trends to be captured automatically, revealing how operational, electrical and mechanical problems and their parameters change with time.

Software used for Condition Monitoring

With sensors recording various parameters of machines as they’re working, it would be handy to use a software application to collect the information and communicate the actions required. The implementation of condition monitoring software is showing rapid growth as manufacturers design efficient and easy ways to interpret data collected by CM systems, and then action it timeously.

One example of this type of software is the Seebo remote condition monitoring solution. Not only does it consolidate the CM data, but it also supports the planning and implementation of a complete CM solution from the ground up.

Condition Monitoring software answers many questions, including where the sensors should be placed, what they would measure, how they should be calibrated, and what alerts should be sent out. This allows stakeholders to get involved in the design of their system at any stage.

Once a remote CM solution has been deployed, the software starts acting as a hub, collecting all the data coming in from the sensors and storing it in a central repository. This allows for corrective action to be driven by deep data analysis.

Condition Monitoring is used for Creating Business Value

Condition Monitoring is in fact merely the first phase in a bigger cycle of doing maintenance through industrial IoT. While the condition of an asset is being monitored, data is collected and stored. If the data determines that immediate action should be taken, e.g. repairs or preventive maintenance, a technician or maintenance team is sent to take action.

Irrespective of which action is taken, the asset’s state together with the sensor data is stored in a data repository. The data repository can be used to conduct comparisons that need historical data, and is also useful to monitor trends and make predictions.

The depth and accuracy of data collected by Condition Monitoring systems, and their reach across entire plants or factories, gives manufacturers very valuable information that could be leveraged to reduce the impact of the Six Big Losses by making informed business decisions.

Trends can be monitored by using Big Data analysis and this forms the foundation for accurate predictions to help both daily operations, as well as triggering proactive and creative strategies for future growth.

The Best Outcome combines Humans and Machines

Condition Monitoring slots into an Industrial IoT framework as a foundational block for continual improvement. Humans must act on insights obtained from CM systems, and the new knowledge must be embedded into new product development, production planning, aftermarket sales and customer service processes. IoT use cases in manufacturing, including Condition Monitoring, can be used to improve day-to-day operations by reducing production and service costs, boosting customer satisfaction and increasing sales.


Quality 4.0 Delivers Improved Performance & Processes to Create Better Products

At first glance, it seems logical that the acceleration of technological growth should automatically bring improved manufacturing quality, but in many cases, the opposite is true. There are a number of factors that make it more difficult than ever before for manufacturers to maintain production quality at a high level.

Production processes nowadays are complex and often encompass many production assets. Whether automotive motors, chocolate wafers, or chemical pest control products are being manufactured, quality failure’s financial impact is higher than ever before.

It is therefore no surprise that quality has become a top priority for many organizations. As Industry 4.0 is making a massive impact on manufacturing, it’s logical that those technologies are being leveraged to meet the new demands on quality. This has resulted in the birth of Quality 4.0.

Quality 4.0 Defined

Quality 4.0, like Industry 4.0, doesn’t simply define a single activity or technology. Quality 4.0 rather describes a new approach to manufacturing that is driven by data. Rather than production being evaluated only on cost and output rate, it is measured by the quality of the product, the process and the services surrounding a product.

The “4.0” refers to Industry 4.0 and the digital transformation technologies associated with it, such as AI in the form of artificial neural networks and machine learning algorithms, Digital Twins, Industrial IoT and others.

All these technologies can be leveraged to improve quality. Predictive Quality Analytics is for example a use case that uses those technologies to predict production quality changes. This information is critical to manufacturers who understand that quality is important to customers, and who want to develop leaner operations while manufacturing better products.

The time for Quality 4.0 is now

Quality 4.0 is still at the early stage of adoption. Most manufacturers still use traditional quality evaluation methods, although these methods are many cases no longer relevant. Companies that don’t innovate on quality for existing and new production processes will find it difficult not only to survive, but also to lead in future markets.

The reality is however that quality issues cost companies an enormous amount of money, and this ultimately affects the potential longevity of any manufacturing concern, especially since the market is always changing and more competitive than ever before.

Quality 4.0 opportunities

Although implementing Quality 4.0 needs organizational and financial resources, the process presents massive opportunities for manufacturers. Finding new, innovative ways to improve quality presents an opportunity to nurture a developmental culture, which could in turn lead to improved products that are less costly to produce.

Implementing Quality 4.0 can also differentiate and strengthen a brand within current markets, and increase awareness, both among existing and new customers.

Quality 4.0 levels the playing field in manufacturing as mid and even small-scale firms are able to leverage new technology, thereby making significant improvements in production efficiency and meeting customer demands better.

Current Quality Challenges

Manufacturers currently face several challenges relating to quality:

  • Allocation of resources for innovation and research into new quality improvement methods.
  • Maintaining high quality levels amidst customer demands changing and higher expectations.
  • More product variety requires agility as work has to be performed on several products at the same time.
  • Complying with changes in laws and regulations.
  • Companies producing from a number of locations have to standardize globally by offering consistent output quality, irrespective of differences in the standards of local production conditions and raw materials.

Industry 4.0 allows manufacturers to meet the challenges mentioned above head-on by using its suite of powerful use cases including digital twin, remote monitoring and predictive quality and maintenance. Changes in legislation can for example be communicated to production lines directly, or code can be remotely changed to ensure that existing and new products comply with the changed laws.

The Four Zones of Adopting Quality 4.0

  1. Concept and Design

Previously, quality was normally associated with production processes. Quality should however also form an integral part of conceptualizing, designing and industrializing of the product.

By including quality in all the phases of the product’s lifecycle, including the early ones, manufacturers will be able to improve customer satisfaction. The quality of a product’s concept is after all an attribute that will ultimately affect how customers experience the use and value of the product.

  1. Production

Before the Industry 4.0 revolution, most of the quality activity in manufacturing took place in this zone. Traditional process harmonization methods and data analytics are now replaced by techniques using machine learning and artificial intelligence, and advanced levels of visualization and monitoring such as Digital Twin.

Rework and production waste is minimized by applying process-based machine learning. Process engineers then test production parameters to determine set points for optimized throughput and quality by using predictive simulation.

  1. Field Servicing and Performance

Quality 4.0 is unique in that a product’s performance is monitored even after it has been delivered. It is also possible to modify the performance if necessary.

Future failures can be prevented, minimizing loss of materials in rejected batches by collecting and interpreting user data from the field. The time taken from identifying failures to eliminating it can be very short, and this reduces wastage and maintains customer satisfaction.

In products integrated with software, updates can be done remotely, to eliminate bugs and add features required by users.

  1. The Company’s Culture

Quality 4.0 is a very wide field and firms should aim at instilling the quality approach as an integral part of the company’s overall culture.

As every interaction with the manufacturing process and every employee can be treated as being within the quality paradigm, Quality 4.0 can be used for more than any specific manufacturing segment.

The Future of Quality with Industry 4.0

Technologies used in smart factories, i.e. Machine Learning, AI, Big Data, IIoT, etc. can all be used for quality improvement. Methods for quality improvement are however not keeping pace with the development of other technologies aimed at enhancing production. This is particularly true for methods that involve feedback loops and B2C communication. This means that the power of Quality 4.0 has not yet been fully unleashed.

Industrial IoT techniques, i.e. analytics, dashboards, gateway devices, sensors and connectivity protocols do however provide the perfect tools for implementing Quality 4.0.

One way to demonstrate this is by looking at remote monitoring as a use case for Quality 4.0.

Quality 4.0 Remote Monitoring

Remote diagnosis techniques can be performed by using sensors to collect data for root cause analysis. When feedback from several devices is gathered, swarm intelligence can be used for further analysis into product performance or machine behavior.

Predictive maintenance can be enabled and correlative patterns identified by using predictive analytics on the data collected. This type of analysis can provide insights into parameters affecting performance quality or output beyond the prevention of malfunction and efficient maintenance.

Although it is commonly believed that only data or software related issues can be fixed remotely, it has become evident that in the field, service technicians are called out for just that reason extremely often. For example, software issues represent a big part of all the service requests in the automotive industry. In cases where on-site visits are required, technicians already know the details of the issue when they arrive, and are equipped with the required methods, tools and spares for the specific repair.

Remote maintenance and monitoring enables manufacturers to improve quality continually over time while the performance and usage data collected from production machines or products provides an important source of information for product development and business insights.