IoT: main challenges and applications

IoT: main challenges and applications

Challenges of IoT and IIoT

  • Data Storage

    Depending on the application, there could be high data acquisition requirements, which in turn lead to high storage requirements;

  • Cybersecurity

    As the Internet of things spreads widely, cyber-attacks are likely to become an increasingly physical (rather than simply virtual) threat;

  • Safety

    A problem specific to IoT systems is that buggy apps, unforeseen bad app interactions, or device/communication failures, can cause unsafe and dangerous physical states, e.g., "unlock the entrance door when no one is at home" or "turn off the heater when the temperature is below 0 degrees Celsius and people are sleeping at night". Detecting flaws that lead to such states, requires a holistic view of installed apps, component devices, their configurations, and more importantly, how they interact;

Applications of IoT

Smart home

Home automation implies the application of smart technologies in a home environment.
Few examples are:

  • Energy control: it is possible to have remote control of all home energy monitors over the internet incorporating a simple and friendly user interface;
  • Light control system: a “smart” network that incorporates communication between various lighting system inputs and outputs, using one or more central computing devices;
  • Using Voice control devices like Amazon Alexa, Google Home or mobile applications to manage coffee machines, ovens, fridge etc.;
  • Connected Household appliance:;


  • Alcon (part of Novartis) has licensed Google’s smart lens technology which involves non-invasive sensors embedded within contact lenses. The lenses may eventually be able to measure glucose levels of diabetes patients via their tears and then store the information in a mobile device, though Novartis backtracked on a plan to test the system in 2016;
  • Some dental insurances provide connected toothbrush to offer special deals to their affiliates by monitoring their daily cleaning behavior.


  • Maintaining Vehicle health
  • Traffic Monitoring
  • Improving Fleet Logistics

Remote monitoring

Big constructions like Dams, Bridges, Buildings, Tunnels and so on can be constantly monitored in real-time.


Amazon introduced computer vision and machine learning inside supermarkets, called Amazon GO. You get access thorugh your mobile phone scanning a QR code, then cameras and algorithms create your virtual cart. No final cashier, simply leave the shop and you will be charged automatically.

Applications of IIoT

Asset Tracking

It refers to the method of tracking physical assets, either by scanning barcode labels attached to the assets or by using tags using GPS, BLE or RFID which broadcast their location. These technologies can also be used for indoor tracking of persons wearing a tag.
Asset inventory:

  • Helps to manage physical capitals and allow to make more informed decisions regarding inventory, such as when to repair or replace items;
  • Maximizes employee and equipment efficiency;
  • Reduces equipment downtime through better planning;
  • Prevents theft and enhance security of items.

Flexible / Reconfigurable Manufacturing Systems

Reconfigurable manufacturing systems (RMSs) are attractive options for handling product personalization, as the system can be continuously reconfigured in accordance with the demanded volumes and products. However, the development of the RMS is a particularly challenging task compared to the development of a traditional manufacturing system (04).

IoT is an enabling technology to achieve RMSs: at BMW plant in Munich, BMW 3 and 4 series are produced, with a throughput of around 900 cars per day. Since this the smallest BMW plant in the world, the lack of space forced to have a single body shop line for all models. The process is 95% automated, ABB and KUKA welding robots are used. Each car body is tracked with a QR code placed in the front: when the body arrive at a welding station, a robot scans the code to recognize the model, and then the process starts.

Welding car body of the all-new BMW 3 series at body shop, BMW Group Plant Munich
Figure 6: Welding car body of the all-new BMW 3 series at body shop, BMW Group Plant Munich . Source:

The same method is applied in the paint shop, which involves 6 different steps:

  • Polishing
  • Phosphate coating
  • Cathodic dip painting
  • Base coat 1
  • Base coat 2
  • Final Paint

All cars flow in a single piece flow line in clean room condition. At the beginning of the line, the QR code is scanned so the right color is identified.

BMW Plant Munich, BMW 3-Series, Paint Shop
Figure 7: BMW Plant Munich, BMW 3-Series, Paint Shop. Source:

Predictive maintenance

Machines are subjected to periodical maintenance to prevent in service failures which can cause a lot of problems in a production system:

  • low productivity
  • higher cost
  • unpredicted repair time, causing longer lead time and in the end lower customer satisfaction

There are different ways to approach Maintenance:

different strategies to approach maintenance
Figure 8: different strategies to approach maintenance. Source: Research Report: Using the IIoT to enhance Predictive Maintenance,, 2018, page 6

Total Productive Maintenance (TPM) is a Lean technique and it focuses on keeping all equipment in top working condition to avoid breakdowns and delays in manufacturing processes. TPM can be considered a Level II approach (Planned).

Today, the combined benefits of the cloud, including lower cost of ownership, high-power computing resources and IoT connectivity capabilities, have enabled more advanced AI-based systems that leverage machine learning to generate higher-value analytics than more basic, model-based systems. Which means, in other words, that it is now possible to predict breakdowns caused machine failures by monitoring the health of key subsystems, such as spindles, electric motors, bearings and so on.

Statistical Process Control

SPC is a widely used Quality tool in Industry to monitor process deviations. The basic idea is to monitor key process variables (KPV) to prevent defects. This mean that data must be recorded and subsequently analyzed. IoT enables automatic solution for Data Collection.

A typical example are smart gauges which can automatically collect measured values. The same principle is applied on smart torque wrenches which record and collect torque an angle values.

Such data can be then automatically analyzed and eventually a real-time feedback (e.g. warning or corrective action) can be provided by AI system.

Mistake proofing

Also called Poka-Yoke, mistake (or error) proofing is a technique used by lean manufacturing to prevent mistakes.

Some examples are:

  • barcode, QR code or RFID instead of human readable methods prevent typos;
  • Smart torque wrenches use RFIDs to automatically set different torque values for different applications;
  • Asset tracking can be used to prevent people from taking wrong parts or assemble parts in the wrong position by tracking the arm:


4) Bejlegaard M et al, Reconfigurable Manufacturing Potential in Small and Medium Enterprises with Low Volume and High Variety, 3rd International Conference on Ramp-up Management (ICRM), Procedia CIRP 51 (2016), 32–37 

Notify of
Inline Feedbacks
View all comments
Would love your thoughts, please comment.x