Embedded systems were historically designed as isolated solutions that performed strictly defined functions inside a closed infrastructure. In the early days, the threat model was mostly limited to physical access and internal configuration mistakes.

But as networking evolved and devices became connected to corporate networks and the wider internet, the risk landscape expanded dramatically. Remote access, cloud services, and integration with external systems turned embedded devices into full participants in the broader cyber environment.

Security in this new reality includes both a technical layer (device, network, infrastructure) and a human layer (processes, behavior, trust, and mistakes).

The Human Factor Didn’t Start With IoT

Long before “smart kettles,” smart bulbs, and modern IoT, hackers were already using social engineering to obtain the information they needed for unauthorized access.

A well-known example is Kevin Mitnick, an American computer security consultant and convicted hacker, who demonstrated that vulnerabilities may exist not in software, but in organizational processes and people’s trust. At the time, it wasn’t always the system itself that was vulnerable, people simply trusted it (and the attacker). Mitnick would call administrators, impersonate legitimate users, and obtain the access he needed.

His work helped illustrate a key idea that still holds today: system protection is not just a piece of hardware between external and internal networks, it’s an ongoing process.

The Internet Exposure Problem: Bigger Than It Looks

Today, a huge number of devices either have direct internet access (sometimes even with a static public IP address), or are connected via intermediary services that interact with them through predefined workflows.

Real-world practice shows that some devices supporting critical infrastructure may have 

• no effective access restrictions
• security left at default settings (for example: admin/admin)

Practical example: public warnings as a routine practice

Services like Shodan highlight the scale of the issue by indexing internet-exposed industrial automation devices.

In Germany, the Federal Office for Information Security (Bundesamt für Sicherheit in der Informationstechnik, BSI) regularly publishes warnings about such incidents, making it clear that this topic is not theoretical. And today, this is more relevant than ever.

Internet-exposed SCADA and energy systems (HMI panels, web interfaces for critical infrastructure, dispatching systems, and more) have long posed and continue to pose a serious threat to public safety and security. One example that explores this risk is the research titled “Investigation of risks for Critical Infrastructures due to the exposure of SCADA systems and industrial controls on the Internet based on the search engine Shodan” by Technische Hochschule Brandenburg.

Vulnerabilities Exist at Both the Physical and Software Levels

To prevent and mitigate threats, organizations use mechanisms to analyze vulnerabilities across multiple layers, including:

• Physical access vectors (e.g., via USB interfaces, CAN bus)
• Software access vectors (e.g., publicly exposed input/output ports, open services, misconfigured interfaces)

In this environment, it’s not enough for a manufacturing company to “build a secure system once.” Security requires continuous monitoring and response. And this is exactly where PSIRT comes in.

What Is PSIRT and Why It’s Not “Just a Team”

PSIRT stands for Product Security Incident Response Team. Despite the name, PSIRT is not merely a group of people responsible for a single product (or product line). In mature organizations, PSIRT is best understood as a repeatable process that includes:

• Monitoring publicly available and subscription-based security bulletins, for example the National Vulnerability Database (NVD) and Common Vulnerabilities and Exposures (CVEs).
• Evaluating how disclosed vulnerabilities affect the product.
• Regularly delivering security updates for deployed systems that the team supports.
• Coordinating communication with customers, clients, and external researchers.
• Publishing security advisories and related documentation.

In practice, PSIRT is not just reacting to CVEs. It is continuous monitoring, risk assessment, fixing, communication, documentation, and improvements to the development process.

PSIRT belongs inside the secure development lifecycle

A key point: PSIRT should be integrated into the product lifecycle, often discussed as Secure SDLC / SSDLC, rather than existing as an isolated function. For an accessible overview of secure development foundations, see OWASP’s guidance on Secure Development.

External testing also matters

A strong practice is also to collaborate with independent specialists who test systems for vulnerabilities from the outside, commonly known as penetration testers.

Why Embedded Products Are Especially Hard to Secure

Embedded systems almost always include a complex stack, for example:

• RTOS or an operating system based on the Linux/BSD kernel
• Third-party libraries (OpenSSL, mbedTLS, zlib, etc.)
• Device drivers
• Proprietary software components
• User interaction interfaces (UI and/or APIs)

Yes, there are important technologies that help, such as:

• Trusted Platform Module (TPM)
• Robust Auto-Update Controller (RAUC)
• Firmware/code signing 
• Rollback mechanisms to return to a previous firmware version

But these are only parts of the overall security system. Without an operational process around them, they won’t deliver their full value.

What a Well-Built PSIRT Process Should Do

A strong PSIRT process should do the following (at minimum):

Long Device Lifecycles Mean Long Security Obligations

Because embedded devices (especially in industrial automation) often have long lifecycles, manufacturers are expected to provide security support and updates over extended periods.

At the same time, even with a mature PSIRT process, security does not depend on the manufacturer alone. The final security posture is also shaped by:

• system configuration
• operational discipline
• timely patch deployment
• organizational controls and procedures

PSIRT, Regulation, and Standards: It’s Becoming Mandatory

PSIRT is increasingly part of broader product security governance.

Cyber Resilience Act in the EU

In the European Union, the Cyber Resilience Act (CRA) requires manufacturers of digital products to ensure cybersecurity throughout the product lifecycle. In effect, the CRA formalizes vulnerability management practices that PSIRT processes have traditionally implemented.

Three CRA Scenarios: What the Cyber Resilience Act Means for Your Business

Industrial automation: IEC 62443

In industrial automation, similar expectations are reflected in the IEC 62443 family of standards. In particular, IEC 62443-4-1 describes requirements for a secure development lifecycle, including vulnerability management processes.

Summing Up

Modern embedded systems are no longer isolated devices operating in closed environments. Once connected directly or indirectly to the internet and external services, they inherit the realities of the modern threat landscape.

A properly designed PSIRT process is not just a compliance checkbox. It is a practical mechanism that reduces cyber risk by enabling continuous vulnerability monitoring, applicability analysis, patching, communication, and lifecycle security improvements.

If your organization is building or operating embedded and IoT products, SaM Solutions can support you in designing, implementing, and maturing a PSIRT capability, from initial process setup to day-to-day vulnerability management.

In the previous article, we examined AI through the lens of John Maynard Keynes and the history of technological change. If we accept that automation does not eliminate work entirely but transforms its nature, a new business question emerges: What happens when technology evolves faster than new markets and roles can form?

For business leaders, this is not a theoretical but a structural issue.

When Productivity Rises but Demand Does Not

With artificial intelligence, the dynamic looks like this:

• Tools are becoming more powerful.
• Companies operate more efficiently.
• But new areas and tasks do not grow at the same pace.

As a result, businesses are making the same product in the same market, but with fewer employees. Work volume decreases. Competition among specialists intensifies.

From a cost perspective, this may look like optimization. From a market perspective, it may signal contraction rather than expansion.

A Business Case Study in AI Disruption: The Stock Illustration Market

A clear example can be found in the stock illustration industry.

Until a certain point, the market thrived on freelance illustrators, stock image platforms, and licensed visual content. With the arrival of generative tools such as OpenAI (DALL-E), Midjourney, and Stability AI, the cost of producing acceptable-quality images dropped sharply.

Companies began generating visuals internally. As a result:

• Demand for mass-market stock illustration declined
• Income for many stock artists decreased
• The broader visual content market survived
• But the low-cost stock segment contracted significantly

The lesson for business is clear: AI reduces production costs, but it may also compress entire segments of value chains.

The Opposite Scenario: When Productivity Creates New Markets

However, cost reduction does not always lead to a decline in employment. There is also the opposite scenario, where increased productivity triggers a new wave of demand for specialists.

For example, the emergence of compilers and high-level programming languages made software development easier and more affordable. This opened the door to mass-market software, web applications, mobile development, and many other fields.

As a result, the number of jobs increased, and entirely new competencies emerged.

Core Competencies in the Age of AI

If AI changes the structure of work, what capabilities become strategically important for professionals and organizations?

Market Expansion or Market Overheating?

In an environment of cheap generation, two macro-scenarios are possible:

1. Expansion with new ideas and directions.
2. Overproduction with excessive solutions and declining differentiation.

A specialist who understands why a product exists and what problem it solves is in a more stable position. AI does not replace professionals. It raises the level of thinking required from them.

From Executor to Structural Shaper

Each of the skills described performs a specific function. 

• Clearly formulating the task reduces the risk of creating solutions that do not initially correspond to the real problem. 
• Checking the result helps to distinguish a convincingly formulated answer from a correct one. 
• Architectural thinking allows you to consider the impact of new solutions on the entire system, not just on a separate fragment of it. 
• A systematic perspective helps you see the interrelationships and consequences of changes. 
• Integration turns the use of AI into a structured process, rather than a set of disparate actions. 
• Responsibility keeps the specialist in the role of the one who makes the final decisions.

These skills do more than just increase personal effectiveness. They determine whether a specialist will remain simply a faster performer or become someone who changes the very structure of work. In the first case, AI reduces the number of participants in the process. In the second, it creates new roles and directions.

Almost a century ago, John Maynard Keynes predicted that technological progress would eventually solve humanity’s problem of scarcity. Instead of working to survive, future generations, he believed, would work to fulfill themselves.

Today, as artificial intelligence begins to automate not just physical labor but cognitive work, including software development, Keynes’s question feels strikingly relevant again. Is AI eliminating jobs, or is it transforming the very structure of work in IT? This article explores how the rise of generative AI echoes past technological revolutions and what it really means for productivity, skills, and the future of the profession.

Keynes’s Prediction: From Scarcity to Abundance

In 1930, John Maynard Keynes, in his essay Economic Possibilities for Our Grandchildren, suggested that humanity’s central economic problem — the struggle for subsistence — was not permanent. Through capital accumulation, scientific progress, and machine-based production, productivity would increase so dramatically that within a century society would face not scarcity, but abundance.

Keynes argued that once productivity reached a sufficient level, work would cease to be primarily a means of survival and would instead become a domain of self-realization and voluntary participation.

The Machine Revolution and Technological Unemployment

Keynes viewed the transition from manual labor to machine-based production as a historical turning point. For the first time, machines systematically replaced human muscle. This led to what he called “technological unemployment” — a temporary imbalance between the speed of automation and society’s ability to create new roles.

However, in the long term, he expected working hours to shrink and the focus of human activity to shift from pure production toward cultural and intellectual development.

The Shift Toward Automating Thought

Today, we are witnessing a similar but deeper transition: from the automation of physical labor to the automation of cognitive labor. Artificial intelligence and “vibe coding” tools are beginning to replace not only manual tasks but also substantial portions of intellectual work, including programming, analysis, and design.

If the industrial era mechanized the body, the AI era mechanizes thinking. This raises the same fundamental question Keynes posed nearly a century ago: if productivity grows faster than the need for human participation, how will employment structures, skill value, and the very nature of work change?

The Paradox of Generative AI in Software Development

The current AI landscape in IT is deeply contradictory. Generative models can produce vast amounts of code, yet that code is often redundant, non-idiomatic, or poorly integrated into project architecture.

Some open-source communities are already facing an influx of pull requests generated fully or partially by large language models (LLMs), requiring more time to review than manual contributions would have required. Many such submissions are rejected, and some projects have even restricted AI-driven “bug hunting.”

In this sense, AI introduces a new form of “technological noise”: quantitative productivity increases, but qualitative output does not always follow.

The Hidden Costs of Replacing Developers with LLMs

In some companies, attempts to replace development teams with LLM-based tools have resulted in hidden costs: growing technical debt, degraded architecture, and more complex maintenance.

Code can be generated quickly. But system understanding, architectural responsibility, and long-term design thinking do not scale at the same linear rate. This fuels anxiety among professionals who see AI not as a tool, but as a destabilizing force in the labor market.

A Recurring Pattern in Technological Revolutions

Viewed historically, what we are experiencing is not a unique crisis but a typical stage of a technological revolution.

Every major technological transformation lowers barriers to entry and increases “surface-level” productivity while simultaneously generating an oversupply of low-quality output. This happened during industrialization, mass production, the rise of personal computers, and the spread of the internet.

The transition period is always marked by overheating expectations, illusions of rapid human replacement, and eventual correction.

AI as a Redistribution of Complexity and a New Infrastructure Layer

Following Keynes’s logic, AI in IT is less about eliminating labor and more about redistributing complexity. Machines begin to automate the routine cognitive layer, while value shifts upward, toward architecture, systems thinking, integration, verification, and risk management.

It is not the disappearance of the profession, but its transformation.

As with previous technological shifts, AI initially triggers anxiety, redistributes roles, and intensifies competition. But in the long run, it expands society’s productive capacity.

In this sense, artificial intelligence can be viewed not as a replacement for specialists, but as a new infrastructure layer, comparable to the steam engine, electricity, or the internet.

The Core Question of Productivity

The fundamental question remains the same as in 1930: will rising productivity reduce routine work and lower burnout, or will it simply intensify expectations and competition among professionals?

The issue is not the technology itself, but whether it enhances human productivity or is used primarily for short-term cost optimization. AI can serve as a cost-cutting mechanism, or it can become an accelerator for research, prototyping, and interdisciplinary integration.

Automation Raises Complexity, It Does Not Eliminate Work

Historically, automation has not destroyed work. It has raised the level of complexity. Machines displaced physical routine labor but created engineering, managerial, and scientific professions.

Similarly, AI automates part of cognitive routine work, freeing space for more complex tasks: systems design, modeling, scientific hypothesis generation, and large-scale architectural solutions.

In this sense, AI may function not as a tool of replacement but as a driver of a new scientific and technological cycle, reducing the time between idea and implementation.

The real question is not whether specialists will disappear, but which competencies will become central in a rapidly evolving professional environment.

(If you prefer video content, please watch the concise video summary of this article below)

When it first emerged on the global IT market, Internet of Things technology caused a sensation. Internet of Things (IoT) solutions have proved to successfully revolutionize the ways that business is conducted across a multitude of various industrial domains on the worldwide level. In this article, we will focus on using IoT in inventory management. Read on to explore it in more detail.

What Is IoT and Inventory Management?

Internet of Things (IoT) represents correlated systems that include smart devices equipped with sensors, software, and other technologies that can receive, process, and transfer data via the internet without human intervention.

Sensors constantly transfer data on the current state of the active devices in real-time. Based on their unique configuration, they can adjust their functionalities according to pre-set requirements. For example, if products stored in a warehouse are overheated, the IoT sensors can send signals to the IoT-based thermostat system in order to cool the temperature down.

IoT inventory management refers to the supervision of numerous processes when the managers order, store, use, transport, and sell the firm’s inventories. All these activities may include work with raw materials, constituent elements, and/or ready-made products.

Impact of IoT on Inventory Management

Increased Efficiency of Supply Chain Management

With the help of IoT-based devices, specialists utilize a number of technologies to collect and manage data in a touchless way. The most widespread technologies include Global Positioning System (GPS) and Radio-frequency identification (RFID).

Such systems are well-equipped with special chips, smart devices, and mobile sensors that carry out effective monitoring and verification activities. Their aim is to ensure that all stages of the process are in order. Thanks to the huge amounts of data kept on IoT devices, the processes in companies have become more transparent, and it’s become easier for managers to track them.

Effective Warehouse Management

The development of IoT-oriented technologies have proven to be really advantageous for the enhancement of warehouse management. In essence, all warehouse management activities are aimed at optimizing storage – creating efficient space allocation and a convenient plan for product storage. For example, products that are rarely used are stored farther than those that are used more often. As a result, this method of space allocation makes accessing them quicker and easier.

Artificial Intelligence Algorithms

With the rise of artificial intelligence (AI) and machine learning, smart algorithms are now able to facilitate all tracking and control activities when dealing with inventory management. These modern technologies can do it better than human beings – they are faster, tireless, and more accurate.

AI is a great tool to discover tendencies on the IT market, as it can effectively analyze huge amounts of data. IoT inventory management specialists can make better decisions based on the results of such thorough research and even protect equipment with such innovative methods as ML-based predictive maintenance. In addition, AI systems can assist in multiple routine tasks at IoT inventory management facilities, such as replacing orders or identifying missing items.

Precise Location Tracking

IoT tools allow more precision while tracking the location of items. If products are equipped with IoT tracking systems, it helps store, move, and deliver them more efficiently. IoT inventory management upgrades the tracking process.

It offers the opportunity to monitor the transfer of products and plan their routes more thoroughly with the help of smart devices. For example, in case of a delay, managers can easily detect when and why the shipping process was disrupted. The analysis of such situations can upgrade the supply chain strategy and find new ways to make the delivery process faster.

Accurate Lead Time Estimation

IoT inventory management allows effective tracking, which is vital when it comes to lead time estimation. With Internet of Things technologies, it becomes possible to calculate the time allocated for final production.

If the assembling activities take a good deal of time, managers should consider time optimization in order to reveal the pitfalls and problems that are to be eliminated. As a result, the operations become more synchronized and effective, which subsequently leads to decreased lead time and driving down costs.

What Are the Benefits of an IoT Approach?

Internet of Things is a revolutionary technology that’s helping to drive the increasingly complex digitization of manufacturing facilities. In the below sections, we will explore the most significant benefits that IoT inventory management may bring to an organization.

Automated Inventory Tracking

RFID instruments spare IoT inventory management specialists from time-consuming routine tasks. In order to track and report on updates, they do not need to do it manually. The IoT system tracks the data and records it to the big data warehouse without human intervention. With such a high level of automation, IoT systems reduce the chance of human-made errors, and save both time and money.

Visible Inventory Characteristics

With IoT technologies, inventory management specialists can ensure that all the processes are highly visible and transparent to all interested parties. You can easily monitor how many items are being produced, stored, or transported in real-time, as well as track their location and state.

When an item leaves a department and moves to the next stage of the production process, this change is instantly reflected in the management system so that the IoT inventory management specialists can see the updates and make relevant decisions if necessary.

Enhanced Inventory Control

It is essential for inventory management specialists to have a clear view of the situation regarding all their products’ location and current status. If they don’t possess such knowledge, it will be hard to find the necessary items when the time comes. If the inventories are in order, it is possible to control the number of available items, avoid a probable shortage, and always meet the demand of the clients.

Bottleneck Analysis

Smart devices provide a thorough description and analysis on the state of all items at a facility. Based on this comprehensive information, managers can unveil all the obstacles that block the manufacturing process and disrupt smooth workflows. With this knowledge, it becomes possible to optimize the processes, making them faster and more efficient.

Optimized Lead Time

As a result of an increased awareness of all the details of inventory such as their number, location, and state, managers are able to make more weighted decisions to improve the production process. Consequently, the lead time is optimized, which leads to a more cost-efficient approach to manufacturing.

Besides, with such groundbreaking innovations as predictive analytics, machine learning algorithms can predict the possible demand and specialists can then adjust the resources accordingly.

How Does an Inventory Management Solution Based on IIoT and RFID Work?

In order to shed more light on the functioning of IoT smart inventory management, we would consider the example of the following industrial manufacturing process. Imagine the manufacturing company has two facilities: one for building the elements for machines, and another for their assembling.

At the very start of such a process, the elements get passive RFID tags. All tags have one-of-a-kind identification numbers which have data about all of the parts. The identification numbers are kept in the associated big data warehouse.

All the elements with a tag undergo several stages of processing at the manufacturing facility. During this process, the RFID reader scans the tags which serves to relay the identification numbers and all the data about the location of elements to cloud storage.

Then, the cloud conveys the analysis of the received data and determines the location and status of every numbered element. If the element is absent, a notification is sent to the user along with the setting of a special status “missing” to the IoT-based inventory management solution. When the system identifies the location of the element, the status is changed back to the initial “in production.”

When the elements are ready, the managers ship them to the assembling area. Prior to departure, specialists scan the elements with RFID to control the number of elements that are sent to the next stage. Then, after the arrival of the elements to the assembling facility, one more RFID reading occurs to make sure that all items were successfully delivered.

Then, the elements are assembled at the manufacturing facility and all ready-made machines get new tags. The old tags on the elements can be either kept or removed depending on the production requirements. When the machines are moved through the departments, each tag is scanned and the data is sent to the cloud for exact location identification.

The next step is shipping to the storehouse. There, tags are scanned again to detect if all the items are in order and there are no absent tags. In this way, the elements are tracked all the way through their manufacturing, assembling, and storage process till they reach their end consumer. Thanks to this constant tracking with the RFID tagging system, manufacturing companies can ensure a high level of production and comprehensive control at every stage of their production process.

Challenges of IoT in Inventory Management

Although IoT offers a multitude of benefits for inventory management, it can bring some challenges and limitations as well. Foremost, as specialists attach RFID tags to the products that are still in progress, it is necessary to guarantee that their presence doesn’t affect manufacturing procedures.

Another problem is the potentially problematic process of scanning, especially with such substances as liquids and metals. When specialists try to do scanning through such barriers, the reading tends to be incorrect. Also, the price of IoT equipment is pretty expensive. Whereas all business expenditures should be feasible to implement them, it’s important to evaluate if the final product is priced high enough to justify the paid expenses.

Some companies are reluctant to implement IoT in their workflows, because they are not sure if it will be safe enough. As any other software solution for businesses, this can be a vulnerable target for cyberattackers. Bearing this in mind, business owners should pay close attention to proper security activities in order to protect their systems. Luckily, there are plenty of ways to ensure a high level of security, such as blockchain-based IoT applications that possess multiple layers of protection.

Logistical activities may be an issue too, as manufacturers experience loss due to spoiled or lost products. However, this is not the problem that only IoT-based inventory management software may face, as all manufacturing facilities suffer from it the same. In order to solve this issue, IoT technology provides tracking and monitoring equipment to safeguard the property.

To perform efficient IoT inventory management, companies need enough data for proper data analysis and strategic planning. With IoT solutions, this becomes possible.

Popular Brands that Have Implemented IoT for Inventory Management

Amazon

Being a huge international tech firm, Amazon pays a lot of attention to eCommerce and retail solutions, cloud, AI, and digital transformation technologies. Warehouse management in the company is based on IoT solutions that include robotics, automated scanning, and transportation. The IoT activities facilitate smooth and stable workflows, and free the workforce to do less routine and more creative tasks.

DHL

DHL is a well-known company in the logistics sector. It specializes in multinational delivery across almost all of the globe. As a renowned brand in the field, the company needs to provide fast, top-quality services to their clients. In order to meet this goal, they enriched their workflows with automated IoT systems that assist in their procedures. Their technologies include drone systems, robotics, and advanced smart devices that make their work more accurate and productive.

Alibaba

Alibaba is a large brand in the eCommerce industry. Being a leading company for online sales in the global market, it needed effective modern solutions to optimize their internal processes and offer first-rate services to their clients. After implementing IoT in their work processes, they excel at trouble-free and quick delivery. This was a result of utilizing automated warehouse systems and IoT-based inventory management.

Why Should I Order IoT Services from SaM Solutions

With decades of experience, SaM Solutions has gained deep expertise in building functional performant IoT projects. We have successfully delivered IoT-based solutions for businesses of various domains across the entire globe. Our expertise allows us to take on especially challenging projects and help companies thrive in their industries with the help of advanced IoT solutions.

Our IoT services include end-to-end custom development of IoT-based solutions, cloud development, consulting and audits, migration and integration, replatforming, optimization, troubleshooting, maintenance, and support.

A good example of one of our IoT-based solutions is the Smart Delivery Box that we created for Sun E Mates. This IoT solution helps to ensure physical safety of parcel shipments, and protects the packages against theft and extreme weather conditions. In close cooperation with the creators behind this idea, we helped the company bring their vision to life.

Thanks to the talented specialists of our company, we deliver modern IoT-based solutions of excellent quality and security. Committed to the idea of project success, we meet and exceed customer expectations. We would be glad to help you with your IoT project. If you are interested to learn more, please feel free to contact our team.

Conclusions

Inventory management can benefit greatly from enhancement with forward-looking IoT technologies. Thanks to such top-notch solutions, it becomes possible to monitor work processes efficiently, collect and transmit data in real time, control the state of the products, and make more weighted decisions. It is a cost-effective way to upgrade the workflows and optimize inventory management, leading to better customer service and increasing an organization’s competitiveness.

(If you prefer video content, please watch the concise video summary of this article below)

To remain competitive in today’s market environment, manufacturing companies must not only produce quality products, but also do this on time so as not to disrupt supply chains and let customers down. Unfortunately, all areas of production face a high risk of equipment failure and unplanned downtime, leading to missed production and delivery times.

However, manufacturers can minimize or totally eliminate unplanned downtime with the help of condition monitoring techniques and IoT technologies. Keep on reading to learn how it works.

Overview

Sudden breakdowns and equipment downtime arise due to the lack of continuous condition monitoring of machines, which makes it impossible to predict and prevent system failures. Many enterprises still work in the old-fashioned way, making decisions reactively, i.e., when the breakdown has already happened. But there are methods allowing you to act proactively, i.e., take action in advance to prevent such a breakdown.

Proactivity is based on data collection and analysis, and it’s the key principle of condition monitoring and predictive maintenance — advanced approaches that are widely implemented in manufacturing enterprises today. These approaches are similar, since both of them monitor the health of machines and spare parts. However, there are some differences.

Condition monitoring

Condition monitoring requires accurate and continuous input data from a variety of sensors and parameters in real time. This practice includes monitoring a range of equipment parameters (temperature, vibration, rotation, etc.), looking for deviations from predetermined control ranges, and informing operators that it’s time to perform maintenance.

Real-time data in combination with digital twins (virtual prototypes of hardware systems) create a transparent vision of the production process. Through this vision, operators can see deviations in production parameters and prevent emergency situations or mitigate adverse effects.

Predictive maintenance

Predictive maintenance concentrates on the early detection of abnormalities by indirect signs such as sudden spikes or unusual combinations of parameters, even if they are within normal operating ranges.

This is possible due to machine learning algorithms that identify possible scenarios of hardware behavior based on the given performance data. This way, predictive maintenance solutions can forecast which equipment will require maintenance in the near future (e.g., in a month). As a result, companies can effectively schedule maintenance activities and avoid unplanned downtime.

According to various research, manufacturing plants generate large amounts of data daily, but only 10% (or even less) of all available industrial data is used by enterprises to improve operational efficiency. This happens mostly because companies either don’t implement condition monitoring systems at all, or don’t have specialized software solutions for proper analysis of collected data. Such missed opportunities may come at a high price, as each hour of downtime may cost hundreds of thousands of dollars.

How Does IoT Condition Monitoring Work?

Condition monitoring is tightly connected with Internet of Things, because the equipment condition data is collected on is done so via a network of sensors and devices installed on this equipment. The data is then processed either by on-premise enterprise software systems such as ERP and MES, or on specialized cloud software platforms. There is a separate direction — industrial IoT (IIoT) used to enhance processes specifically in plants and factories.

To create a condition monitoring system for manufacturing hardware, you should solve a range of tasks:

• Equip machines on the factory floor with sensors, trackers, controllers, or other monitoring devices
• Set up cloud or on-premise software for collecting, transmitting, and storing data
• Create digital models of equipment operation
• Develop algorithms for data analysis and forecasting
• Set threshold values for parameters and response rules
• Develop reports and data display panels
• Assign responsible employees and train them to work with the system

When all the stages are completed, you’ll get a comprehensive condition monitoring system that will control your manufacturing machines day and night.

Basic hardware parameters for condition monitoring

Condition monitoring systems measure various equipment parameters, which deviations from can indicate coming failures.

• Machine oils and lubricants signalize about overheating or contamination.
• Vibration helps detect wear of spare parts, their misalignment, or imbalance.
• Temperature deviations of mechanical parts may point to wear, lack or absence of lubrication, misaligned spare parts, or foreign objects in machines. Abnormal temperatures in electrical equipment may reflect some leaks, or issues with pressure and current.
• Pressure shows the condition of gas, water, or steam in machines. Pressure deviations warn about leaks or other disruptions of the system’s integrity.
• Rotation frequency is also measured to detect the wear or imbalance of spare parts.

Data stream processing

Internet of Things (IoT) platforms offer ready-made tools and solutions that allow you to quickly create and implement systems for hardware condition monitoring at minimal cost. As a rule, IoT platforms for manufacturing are hardware-agnostic, so you can easily integrate them with various controllers, sensors, and other tracking devices with the appropriate resources.

Data collected by these devices is transmitted to data analysis software tools and then visualized in the form of diagrams, charts, and reports convenient for end users. Such data analysis platforms can be either on-premise or cloud-based.

However, manufacturing machines may generate large data streams with high frequency. And in many cases, it is inefficient and expensive in terms of communication channel bandwidth to transfer raw data directly to the IoT platform.

To solve such issues, there is an edge computing approach, i.e., on-site data processing using gateways and smart sensor nodes. The pre-processed and compressed data is then transferred to a cloud platform or an enterprise system for further analysis.

If the system detects some deviation, it sends a notification/alert message to the responsible employee, who can take instant measures.

How Can IoT Help With Condition Monitoring?

The Internet of Things concept in its essence implies the creation of a network of sensors and trackers installed on various physical objects. Such sensors and trackers are able to collect data, communicate with each other, and transmit data to software platforms for processing and analysis.

There are several reasons why IoT solutions are crucial for condition monitoring of industrial equipment.

Established communication channels

A variety of IoT communication protocols are used to connect hardware machines and ensure their seamless interaction and data transmission. Reliable communication channels are highly important for condition monitoring, as they ensure the timely receiving of information and prompt decision-making.

Remote condition monitoring and management

With IoT technologies, you can monitor and manage various types of equipment remotely. This is essential for certain industries where machinery works in remote terrains, e.g., oil and gas production.

IoT devices eliminate the necessity for maintenance employees to be on-site all the time, as they collect and transmit data on equipment conditions automatically in the online mode. In case of issues, operators can either switch operating modes or stop the equipment remotely, thus preventing failures. This way, companies can significantly reduce maintenance costs because maintenance crews go to the field only when repairs are actually needed.

Cloud computing

IoT offers limitless opportunities for cloud data processing and storage. Taking into account that industrial machines generate tons of data, it’s more cost-efficient and convenient to store this data in the cloud, on remote servers. Moreover, cloud IoT platforms provide enterprise employees and managers with secure access to condition monitoring data from anywhere, which facilitates decision-making.

Machine learning

Modern IoT platforms are equipped with advanced analytics, including AI and machine learning algorithms. This means that collected data can be processed more efficiently and valuable insights on the condition of machinery can be extracted.

Automation

The usage of IoT in condition monitoring results in a high level of automation. Staff employees don’t have to check and test all the machines manually in order to detect malfunctions, since IoT trackers and sensors do it automatically without days off and lunch breaks.

Examples of Condition Monitoring Applications

In fact, all industries that work with some kind of equipment and hardware devices can make use of IoT-based condition monitoring. The following are the most popular fields of application so far.

Renewable energy

As more countries and companies strive to abandon fossil fuels and implement renewable energy solutions, more complicated innovative machinery is put into operation globally. Undoubtedly, equipment for wind, solar, and hydroelectric energy systems require continuous condition monitoring, which can be ensured only with the help of IoT solutions.

Downtime and failures of such systems lead to tremendous economic losses and compromise social spheres that consume energy. To avoid significant accidents, governments and green energy business owners actively invest in IoT condition monitoring systems.

Oil and gas exploration and development

The reliability of equipment (turbines, pumps, compressors, generators, etc.) at oil and gas fields impacts not only the financial health of companies, but also the safety of the environment. Considering that most oil and gas sites are located either offshore or in remote areas with tough weather conditions, using IoT-based condition monitoring is a must for the industry.

IOT MACHINE_LEARNING EDGE

A Prototype for the State Analysis of an Electric Motor

Manufacturing

Both discrete manufacturing (the production of distinct items such as automobiles, household appliances, furniture, etc.) and process manufacturing (the production of foodstuff, chemical and biotechnological goods, etc.) depend on the performance of machinery. All kinds of welding, milling, drilling, rolling and other machines should be continuously monitored to avoid the negative impact of some defects on the entire batch of products coming off the assembly line. Hence, plants and factories willingly introduce IoT condition monitoring solutions.

Transportation

The transportation industry strongly depends on the technical and operating condition of vehicles. Any spare part defect or its amortization may cause a road accident, resulting in delivery delays, significant car damages, and even human victims. IoT technologies and condition monitoring are a go-to for transport companies striving to minimize accidents, reduce delays, and lower maintenance costs.

Ready-Made Solutions for Condition Monitoring by STMicroelectronics

With the advancement of modern technologies and a variety of hardware providers on the market, companies don’t need to invent the wheel and develop condition monitoring or predictive maintenance systems from the ground up. Fortunately, there are ready-made hardware and software solutions for these purposes.

A great example is STMicroelectronics, one of the world’s leading semiconductor companies headquartered in Switzerland. They offer a wide range of energy-efficient products and solutions powered by mobile and IoT devices for smart factories and other industries dealing with electronics.

ST’s portfolio for condition monitoring and predictive maintenance includes microcontrollers and microprocessors, sensors and inertial measurement units (IMUs), as well as wired and wireless communication solutions. For instance, you can quickly and easily develop a data logging and anomaly detection solution with the help of the following products by ST:

• STWIN SensorTile — a development kit and reference design based on an ultra-low-power microcontroller and comprising industrial-grade sensors (accelerometer, vibrometer, etc.). Using it, you can create and test IIoT prototypes.
• FP-SNS-DATALOG1 — a comprehensive solution for data collection and storage from any type of sensor.
• FP-AI-NANOEDG1 — a function pack for creating Machine Learning libraries for each specific project and integrating them on physical nodes. Its distinctive feature is that a NanoEdge AI library can learn normal patterns directly on the microcontroller, without any connection to the cloud.

In other words, if you need to build a condition monitoring system for your industrial equipment, you are free to choose from ready-made hardware and software solutions, combine them to meet your project requirements, and benefit from results — simple as that.

Advantages of Condition Monitoring Systems

The key objective of condition monitoring systems is to help companies avoid potential equipment failures and damages. If this objective is achieved, companies enjoy a number of valuable benefits from using such systems.

• Reduced maintenance costs — doing preventive maintenance is cheaper than fixing serious equipment damages and coping with the hidden expenses of downtime. Moreover, manufacturers can also reduce energy costs due to monitoring critical parameters and ensuring the proper operating mode of hardware.
• Maximized productivity — when all the machines work like a clock and the process is not interrupted by technical issues, employees fulfill their tasks on schedule, hence showing consistently good output.
• Longer lifespan of the equipment — timely maintenance and avoidance of significant breakdowns ensured by IoT condition monitoring prolong the service life of machines, eliminating unnecessary expenses.
• Guaranteed operational safety — in many industries such as chemical, nuclear, biological, transport, and others, machinery malfunctions may cause serious environmental disasters, harming people and nearby territories. IoT technologies and condition monitoring systems prevent machinery from causing catastrophes.

A Success Story

SaM Solutions in partnership with Toradex, a provider of embedded computing solutions from Switzerland, developed a prototype of a predictive maintenance system that monitors motor vibration frequency and force to detect abnormalities, signaling failures. The system can be applied to any industrial equipment with electric motors.

This condition monitoring system collects data on the motor performance via a Toradex MPU-6050 sensor and analyzes it with the help of machine learning algorithms written in Python and based on the Amazon Greengrass cloud service and Amazon Lambda function.

The efficiency of prediction results of this prototype turned out to be up to 99%.

Summing Up

Solutions and tools for condition monitoring and predictive maintenance of industrial equipment are being actively developed all over the world, and IoT platforms play a key role in this process. To remain competitive, increase production transparency, and enhance system uptime, industrial enterprises need to implement IoT and condition monitoring technologies.

SaM Solutions has the skill set to design, develop and implement IoT and embedded projects for monitoring the condition of equipment across various industries. Contact us if you want to get a consultation and learn the details about our related projects.