PostNL has started fitting trackers so that the logistics chain can be managed real-time. It will be possible to track the movement of 250,000 roll containers used for collecting and transporting parcels. These trackers are known as beacons. The Internet of Things (IoT) is an important component in the digitisation of the logistics process, according to Selçuk Sasoglu and Sander Heije.
Selçuk: We are doing more and more with IoT so that we can get a better picture of the total logistics flow. Tracking roll containers in real-time and combining this with data from other parts of the network extensively optimises the process. For instance, by creating better plans based on real-time and historic data. It is really all about the digitalisation of logistics.
Sander: When you also know which parcels you have placed on which roll container, you have a full overview of the logistics process. The container can be followed in real-time, so you can tell customers and consumers exactly where their parcels are. We can step in when a filled roll container is taking too long in a certain place. In this way, we can prevent delays building up somewhere.
If we know that we always get a relatively high volume of parcels on Monday from a certain online store, then we want to make sure that they also have a sufficient number of roll containers. By looking at the actual demand and capacity, we can use calculations to automatically increase customers’ stocks if it turns out that they do not have enough roll containers. Gateway data
Selçuk: We process IoT data on an overwhelming scale. The platform can process millions of messages per minute. It means that it is possible for us to determine the position of a roll container every millisecondwith a very high accuracy. The trade off is we need to process more messages in order to have more accurate detections. More messages mean more costs to run our platform. That is why we continuously weigh how much data we need, how much it will cost and the ultimate return for the business.
A parcel or container passes various types of gateways in the process. It means that each control port provides a specific kind of information. For instance, some determine the location, while other scanners process information about the parcels.
It is more useful if we bring together more detections from various gateways in the process over a slightly longer period. After we have filter out any potentially unreliable data, we can then determine the location or status with a far higher accuracy. Filtering our unreliable data also means that the cloud infrastructure needs to process less data, which lowers the costs.
Sander: IoT technology is of fundamental importance for the digitisation of logistics. As the Internet of Things Platform Owner, I play a role in developing a vision and strategy about the IoT platform, creating teams and managing the budgets. Internally, I try to inspire the organisation by using a proof of concept to show what IoT can deliver for PostNL.
As with the roll container project, we ultimately create a working and scalable IoT platform. In doing this, our ambition is to make as much use as possible of serverless computing, where we pay per use in the cloud.
Selçuk: As the Lead Engineer of the IoT team, I search for a balance between quality, value for the business, time and costs. We try to limit technical debt as much as possible with a view of the long term. It takes more time to create long-lasting and robust software, but it makes you faster and more agile further along the line.
I also help to recruit our engineers, as we would like to build up our own capacity in the area of IoT. If you have some knowledge about AWS, and you enjoy complex challenges, then now is the perfect time to join PostNL.
It is also interesting how we use bleeding edge technology from AWS, such as Location Services, which enables location data to be added to applications. It is still running in a preview phase, but we can already use it.
Sander: Connecting roll containers is the first step. The next phase is connecting other devices, such as bicycles, temperature-controlled containers, cars, etc. For instance, sensors can monitor for jolts or monitor the temperature. This can be used to develop services that add value.
We are also looking at connecting other technologies, such as computer vision cameras combined with artificial intelligence. By combining this information in the IoT platform, we can monitor exactly what is happening in the process.
We are leading the way with the large-scale connection of sensors. Whereas at first it was mainly a technology push, after the implementation of the first project, new ideas and questions from the business emerge everyday.
This is only the first step! If you would like to know more, get in touch with us.