What is IoT and what is it for? The Internet of Things in production and warehousing

What is IoT and what is it for? The path from sensor to decision in production and warehousing, use cases and a simple checklist to start with.

FUFerhat UğurluYazSis TeamPublished: November 20, 20268 min read
In short

IoT (the Internet of Things) means that physical objects produce data through a sensor or tag, the data travels over a network to software, and the result becomes a visible or automated decision. What IoT is for: turning a physical event into digital data at its source and speeding up decisions. In production it shows machine and product movement live; in the warehouse it shows stock and counts live.

In a fictional factory, the production manager is asked a simple question at the month-end meeting: what time did the line stop last night, and which work order was waiting? The answer is in the operator’s shift notebook, and the handwriting is unreadable. The meeting turns into a debate about why the figure arrived late. The problem is not that the information does not exist, but that it reaches nobody at the moment it is created. What we call the Internet of Things closes exactly this gap: it turns something that happens in the physical world into an event the software knows about, at the moment it happens.

IoT, the Internet of Things, means that physical objects produce data through a sensor, tag or similar device, that this data is passed over a network to software, and that the result is used on a screen, in a report or in an automated decision. This article explains the concept with a simple model, shows use cases in production and the warehouse with examples, and describes where you can start.

How does IoT work? A simple four-layer model

Every IoT solution, whatever device is used, consists of four layers. Knowing at which layer the problem lies makes it easier to understand why a project “does not work.”

LayerIts jobExample from production and the warehouse
SensingMeasures or recognizes the physical eventA sensor, an RFID tag, a counter
ConnectivityCarries the data over a wired or wireless networkA wireless network, a wired connection, a handheld terminal’s network
Data processingCleans raw data, interprets it with rules and, if needed, decides on the spotAn edge computer, duplicate removal, threshold checks
ApplicationPresents the result to a person or another systemA live screen, a report, an ERP or warehouse software

The “edge” concept in the third layer is often confused. Edge means that data is pre-processed where the event happens, on a computer next to the reader or the machine, before it goes to the center. Noise such as the same event arriving many times a second, repeats of a signal or a temporary error is filtered out here; only the meaningful result goes to the center. Which decision is made on the spot and which at the center is weighed separately in each project.

What are the use cases of IoT?

The use case is “every physical event that can be measured.” These are common examples in production and the warehouse; all of them are general information and they do not create the same value in every business.

  • Machine status: Watching live whether the line is running, stopped or on standby.
  • Product and work order movement: Recording which station a tagged part passed through and when.
  • Stock and counting: Counting cartons, pallets or hanging items in bulk and quickly.
  • Environment monitoring: Tracking conditions such as temperature and humidity in areas where sensitive products are stored.
  • Asset and equipment tracking: Knowing the location and history of a vehicle, crate or tool.
  • Energy and consumables: Measuring how much resource a line or machine uses.

A fictional example: pallet counting before shipment

In a warehouse, counting pallets before shipment used to be a job done by two people with a list and a barcode reader. When a tagged pallet passes through the read zone, the reader collects the tags, the edge computer filters out the repeats, compares them with the order list and shows the result on screen as “matched, missing, excess.” When the operator approves, the result is passed to the warehouse or ERP software. Here the four layers of IoT show themselves: the tag senses, the network carries, the edge processes, the software decides.

What is industrial IoT?

Industrial IoT is the same idea in work environments such as factories, warehouses, energy and logistics. What sets it apart from consumer uses such as the smart home is the high cost of failure: a line stopping or wrong stock information directly affects production and shipping. That is why in industrial installations reliability, data accuracy, network security and working together with the existing ERP or production software become bigger issues than the device itself. In practice this means moving in the order “first the decision, then the data that will feed that decision,” rather than “first the device, then we will use it somewhere.”

How do you design an IoT sensor production tracking system?

Sensor data on its own is not a decision; it is a number that reaches nobody. In a production tracking design, each of the following steps must be answered.

  1. What will you measure? The machine’s status, the product’s movement, or an environmental condition?
  2. How often? Will you record every event, or a summary at set intervals?
  3. Which threshold is meaningful? Who defines the line between normal and abnormal?
  4. Who will be told? In an abnormal situation, who gets the alert and through which channel?
  5. Which work record will it be linked to? If a measurement is not tied to a work order, a batch or a machine, the report lacks context.
  6. Where will the data go? An ERP, production software or a custom application; the target must be clear from the start.

We should also add the point where IoT alone is not enough: collecting data is one thing, tying data to a work record is another. Every value coming from a sensor or tag has to be linked to a work order, a batch or a location; otherwise the management screen has plenty of numbers but cannot answer any question. For this reason IoT projects are usually considered together with an ERP or production software.

Running it with a notebook, a spreadsheet and shift notes

In a small workshop, keeping machine status in a notebook, counting by hand, or reading the temperature a few times a day and writing it into a table still works. The method is not expensive and everyone understands it. Problems show up clearly in three places: data arrives late and by the time it is interpreted it is too late, the record depends on a person and can be skipped, and on different shifts the same thing is written differently. The real contribution of IoT is in these three points: it reduces delay, skipping and inconsistency.

Questions to ask before starting an IoT project

  • Which information arrives late or wrong today, and what does that cost?
  • Does this information come from a physical event, or is it already being entered into software?
  • Will measurement happen at a single point or at many?
  • Can the existing ERP or production software accept this data?
  • Who will decide what from the result? Is that decision being made today?
  • In which small flow will we run the first trial, and by what criterion will we call it a success?

Buying devices before the answers to these questions are clear on paper is the reason most IoT ventures end up in trouble. When there is a clear question, a measurable flow and a decision owner who will use the data, even a small trial shows value quickly; otherwise even the most advanced sensor produces a screen nobody opens.

On the YazSis side: RFID Tunnel, handheld terminals and live data transfer

Because IoT is a wide umbrella, the scope of YazSis needs to be stated clearly. The solution described on the RFID & IoT systems page focuses on the RFID Tunnel, RFID handheld terminals, production tracking software, an edge computer and live data transfer: bulk reading of tagged cartons, hangers, pallets and production parts, seeing the result live on a laptop or terminal screen, and passing the data to a computer, a company server or a cloud infrastructure. On the production tracking side, work order, station, operation, product and time information is matched with RFID movements.

If you would like to work out together which physical event in your business is recorded late or wrongly and start from a small flow, the digital transformation and RFID & IoT systems pages are a good starting point.

Frequently asked questions

Are IoT and RFID the same thing?

No. RFID is a technology that identifies an object by a tag and can be used in the sensing layer of an IoT solution. IoT is a broader concept that covers the sensing, connectivity, data processing and application layers together.

What is the difference between industrial IoT and consumer IoT?

Industrial IoT works in environments such as factories, warehouses and logistics; reliability, data accuracy, network security and integration with existing enterprise systems matter much more. In consumer IoT this burden is usually lighter.

Does IoT necessarily need a sensor?

No. The sensing layer can be a sensor, an RFID tag, a counter or another device. What matters is that the physical event is turned into data automatically.

Where is IoT data sent?

The target depends on the need: a computer, a company server, a cloud infrastructure, an ERP or custom software. What matters is that the target and the form in which the data will go are clear at the start of the project.

Where should an IoT project start?

It is reasonable to pick a single physical event that is recorded late or wrongly today and try it in a small flow. If the result is good, the scope is widened; this keeps both cost and risk under control.

FU
Ferhat Uğurlu · YazSis Team

Builds enterprise software, AI and digital transformation products together with the YazSis team.