Nowadays, there is a gamut of innovative technologies that manufacturing companies can use to enhance operational efficiency. Artificial intelligence, machine learning, 5G, IoT, various automation solutions, and cloud platforms are poised to change manufacturers’ business models for good. However, IoT lies at the core of Industry 4.0 and is poised to generate the most value in the manufacturing context. According to the recent report by McKinsey, IoT has the largest economic potential in factory settings and can generate up to $3.3 trillion by 2030.
Many manufacturing companies turn to IoT software development because it allows them to finally gather and analyze data that industrial machinery produces. By having a greater insight into a factory’s performance, manufacturers can achieve never-before-seen levels of automation, operational efficiency, and productivity. IoT also allows manufacturing organizations to significantly enhance their equipment and energy management systems, and drastically minimize the risks of accidents and machinery failure.
Ignoring these technological advancements and sticking to conventional practices will inevitably lead to comparatively weak operational infrastructures that won’t withstand competition in the long term. While there is an abundance of IoT use cases, manufacturing companies are struggling to realize their value and scale their impact. In this article, we will discuss why now is the time to start exploring IoT possibilities in manufacturing, discover its most potent use cases, and figure out an implementation roadmap.
Manufacturing IoT adoption: now is the right time
In the past few years, we’ve seen significant technological leaps, especially in the field of scalable device connectivity and integration. Tapping into these advancements, manufacturing companies can scale their operations by enhancing existing operational infrastructure rather than completely revamping it.
Furthermore, similar to IoT in telecom, the entry barriers for adopting human-machine interfaces and IoT in manufacturing have become lower, enabling manufacturers to adopt IoT with considerably fewer resources. First, the emergence of low-code development platforms has made complex IoT solutions much easier to create and install while allowing more flexibility at the production floor. Also, given that low-code IoT development requires business users’ and technicians’ involvement, the workforce will be more willing to accept changes caused by IoT transformation.
Second, the proliferation of edge computing enables manufacturing companies to employ IoT without allocating too much computational power and employ a wider range of features.
Third, improved connectivity provided by 5G and next-gen integration frameworks facilitate much faster device communication without compromising security. For example, the previous generation of IoT-enabled automated carts, forklifts, and other vehicles operating on the production floor relied on Wi-Fi for connectivity, which allowed vehicles to drive only pre-programmed routes. With 5G in place, employees can manage vehicles’ tasks in real-time, enabling more flexibility and operational efficiency.
Best IoT use cases in manufacturing
Let’s discuss how manufacturing companies can leverage IoT to streamline equipment maintenance, energy management, and other manufacturing processes.
Predictive maintenance
To fully grasp the value of IoT-enabled predictive maintenance in manufacturing, let’s go over the traditional approach to equipment maintenance. Conventionally, manufacturing technicians replace certain parts of the machinery at regular intervals to minimize the risk of failure. The problem with this approach is that equipment age isn’t a reliable indicator of its condition. In essence, manufacturing companies lose money each time they replace machinery or parts which could have functioned for much longer.
In contrast, the predictive approach to equipment failure is far more efficient and cost-effective. By utilizing smart sensors similar to those used in smart offices, edge computing, advanced machine learning algorithms, and computer vision in manufacturing, industrial companies can accurately predict equipment failure and make real-time adjustments to ensure that production flow is uninterrupted.
So, how does this work exactly? Installed sensors capture data from machinery motors, drives, and other components, then send it to the plant’s cloud or on-premise servers. Afterward, this data is structured in a big data warehouse and analyzed by ML engines to detect performance deviations. It’s crucial to note that besides sensor data (for example, temperature and vibration), the ML algorithm also has to take into account other relevant data like machinery model, configuration, and usage history for higher result accuracy.
One of the most appealing features of predictive maintenance is that the underlying ML algorithm is going to get more accurate over time. This is especially relevant in the manufacturing context, as technicians often struggle to take into account variable environment-specific factors. Overall, ML-powered manufacturing tools are essential for reaping the full benefits of IoT.
ABB and Tenaris case study
Tenaris, a global manufacturing organization that produces steel pipes, has joined forces with ABB, a multinational producer of automated electrical equipment, to streamline maintenance of more that 400 electric motors that drive rolling mills at their Italian plant.
ABB installed smart sensors on each of the plants’ motors that send performance data to a dedicated platform so that the Tenaris engineering team could monitor the condition of all motors in real-time. On top of that, when certain metrics go above the established threshold, the system automatically sends notifications to the maintenance team. Ettore Martinelli, Tenaris Maintenance Engineering Director, claims that ABB’s solution does a perfect job at detecting excessive motor vibrations, which is often a sign of a failure. Similarly, smart sensors have proven to be effective in detecting voltage anomalies that indicate the likelihood of a short circuit.

Sensor-free predictive maintenance
To further highlight the fast pace of IoT technological advancements, let’s discuss the sensor-free predictive maintenance system developed by researchers from the Imperial College London. The alternative to IoT sensors dubbed CogniSense can monitor machinery that exhibits cylindrical motion by transmitting radio-frequency signals and capturing the equipment’s response for further analysis.
While conventional IoT sensors can measure the performance of a single piece of equipment, meaning that an average plant needs thousands of them to roll out a predictive maintenance program, the CogniSense gadget can monitor multiple devices at the same time.
Energy management
Energy management encompasses a set of tools and methods which help manufacturers to efficiently procure, distribute, and use energy to save costs and minimize negative ecological impact. The majority of manufacturing companies have an energy management system in place encompassing meters that gather data to be further analyzed by human employees. This is a traditional, yet inherently cumbersome and ineffective approach, since it doesn’t allow manufacturing companies to adjust energy systems in real-time. Moreover, this pen-and-pencil method is notoriously error-prone.
With the help of an IoT-enabled energy management system, it’s possible to have a comprehensive view of energy consumption per unit equipment or plant at all times. IoT allows manufacturing companies to save costs by reducing transformer losses and standby loads and eliminating the possibility of peak loads. What is more, an automated monitoring system helps manufacturing organizations to streamline energy audits as data collection and consequential reporting is an additional benefit of an IoT-enabled energy management system.
An IoT-enabled energy system works similarly to the aforementioned predictive maintenance solutions. Smart energy meters, installed across the plant, collect various energy consumption information including voltage data, KWH, etc. and send it to the IoT platform. In many cases, especially when it comes to SMEs, this holistic approach to data aggregation is enough for a company to improve their energy management. In other words, complex ML-powered analytics systems are not mandatory to get a return on investment for comparatively smaller facilities.
By understanding exactly how much energy a specific energy consumer needs at a particular time, energy managers can identify consumption patterns. For example, if energy use is unreasonably high at some point, it’s likely there is an unknown energy consumer onsite.
Additionally, the installed sensors can be combined with anomaly detection tools powered by machine learning to identify power consumption fluctuations that may indicate the probability of equipment failure. This is a perfect example of how beneficial a holistic approach to IoT transformation is as IoT-enabled energy solutions can work in tandem with the predictive maintenance system.
Digital twin
A digital twin is a virtual copy of a physical element in the manufacturing process digitally mirroring equipment’s condition, functionality, and connection to other devices.
In essence, a digital twin in manufacturing is a virtual representation of what is happening inside the factory. A myriad of sensors installed across the facility collect product characteristics data including asset color, thickness, temperature, etc. Product engineers can examine these parameters, adjust production systems, and optimize assembly in near real-time, which can lead to cost reduction, productivity increase, and product quality improvement. Moreover, depending on the product type, manufacturing companies can also enhance product performance even when it leaves the production facility, significantly boosting customer loyalty and satisfaction. This can be achieved by remotely analyzing product performance data and tuning parameters to achieve optimal efficiency in near-real time.
Unilever case study
In 2019, Unilever partnered with Microsoft to develop a digital twin of its production facilities. The majority of machinery in Unilever factories is now constantly sending data to an IoT platform providing engineers with much better control over operations. For example, in one instance, the digital twin is using data to assess how long it takes to produce a bottle of shampoo, which helps engineers to optimize the manufacturing process. Similarly, the digital twin now helps control the moisture levels in Unilever’s soap-making machine, enabling consistent product quality. After adopting a digital twin in its facility in Brazil, Unilever saves $2.8 million yearly due to reduced energy consumption and increased productivity.
Manufacturing IoT implementation roadmap
As the examples cited above demonstrate, IoT solutions enable manufacturing organizations to enhance their competitiveness, increase productivity and efficiency, boost revenues, and stay relevant in this digital landscape. While the majority of manufacturing companies understand the importance of IoT, the overall adoption pace is slower than expected. Here are the essential steps that companies need to take to implement IoT and benefit from this transformation.
1. Use case prioritization
One of the most critical things that manufacturing organizations need to realize is that IoT can prove truly efficient only when multiple use cases are implemented. In other words, it’s better to implement many IoT tools than pick the most potent technology and focus solely on it.
Stakeholders should start with drawing up a list of possible IoT use cases. One of the most effective ways to determine promising use cases is to figure out what challenges manufacturing personnel dealt with when completing their tasks. Alternatively, organizations can select use cases based on desired KPIs aligned with their long-term business needs and goals.
After that, it all comes down to prioritizing use cases according to their impact and applicability. You can determine the impact by the cost reduction or profit growth potential of the use case. The second most important metric should be the complexity of implementation, or how many additional resources or how much training is needed to bring the use case to life. Additionally, you can estimate the complexity of implementing a particular use case across your other manufacturing locations. In general, it’s recommended to focus on use cases that can provide immediate benefits.
2. Use cases implementation
When the list of IoT use cases is complete, it’s time for their implementation. Manufacturers need to realize that it’s not mandatory to have a perfectly mature infrastructure to begin the rollout. Regardless, it’s important to establish automated data collection frameworks for a specific use case from the very beginning, which may involve the addition of new sensors, existing software updates, etc. Essentially, as you are rolling out the first few IoT use cases, it becomes easier to finalize the required architecture for an IoT-enabled plant.
3. Organizational change management
One of the most common traps that manufacturing organizations fall into is treating IoT digital transformation as a solely IT project. The implementation of the first use cases implies a rather serious working process transformation for many employees and, most importantly, a shift in their attitude to work. This is why change management should be among the top preparatory measures when embarking on the IoT transformation journey.
For manufacturing companies embarking on IoT digital transformation, reskilling and attracting new talent is an essential success factor. More often than not, when examining the selected IoT use cases in detail, manufacturers discover that they need to fill new roles including business translators, data engineers, project managers, blockchain experts (in case you implement blockchain for IoT security) and provide their existing employees with substantial training programs. According to McKinsey estimates, typical IoT transformation requires up to 90% of the workforce to upgrade or learn completely new skills. It’s also paramount to clearly explain the incoming changes and their benefits to plant workers.
All these steps will help manufacturers with achieving workforce buy-in and increase trust in IoT initiatives.
The dedicated IoT transformation team should make a list of all the skills and knowledge needed to operate new or updated tools. Also, this team should monitor the rollout and be on the lookout for any problems that might occur along the way. This is extremely important as regardless of the plant’s technological maturity, force majeures are likely to happen. On top of that, documenting the implementation process also helps in avoiding similar hurdles down the line.
It’s also a good practice to involve employees that will be operating new tools and applications in the IoT implementation process. Not only will their feedback be often priceless, but their involvement will also be beneficial for the overall acceptance of the transformation. It’s hard to stress enough the importance of communication and collaboration when implementing IoT applications, because way before the transformation brings positive results, it directly affects people that operate those tools.
4. Agile transformation
For the majority of manufacturing organizations, any digital transformation initiative is a sign of a new beginning. Understandably, it’s very tempting to ensure 100% readiness in terms of technology, talent, resources, etc. But in fact, it’s far more feasible and effective to approach transformation head-on, failing and learning as you go. Not to say that you don’t need to prepare at all. Instead, the preparation should center around establishing a dedicated transformation team which utilizes proven transformation management practices.
What is more, in the context of IoT transformation which, like all digital transformations, is a never-ending journey, an agile approach that encourages quick development and continuous optimization is the best choice.
Many executives also assume that IoT implementation requires the elimination of all legacy equipment since their old manufacturing facilities aren’t suitable for IoT. Most likely , this comes from the way the media often portraits industrial IoT as completely automated and remote-controlled facilities without a single human employee onsite. In reality, IoT transformation value comes from enhancing and optimizing existing plants’ infrastructure by continuously adding more sensors, apps, and machinery.
Conclusion
Currently, the IoT is the core technology that can consolidate digital and physical elements in one place, enabling a previously unattainable level of control over various manufacturing processes.
In essence, the success of IoT in manufacturing comes down to putting equal effort and resources into both technology and business-related implementation factors. In the end, manufacturing digital transformation powered by IoT implies rethinking existing operations and business models to streamline workflows and enable data-based decision making rather than deploying new technology solely to cut costs.