web analytics

The Industrial IOTA Lab Aachen

The Industrial IOTA Lab Aachen

Real world adoption of the Tangle technology is happening at an increasing speed as IOTA works its way into industrial and academic circles. The Industrial IOTA Lab Aachen (IILA) at the Laboratory for Machine Tools and Product Engineering (WZL) of the RWTH Aachen University happens to connect both of them. Therefore, it’s a great example worth exploring in more depth.

As an open laboratory, IILA invites students, lecturers, hobbyists, and industry experts to collaborate on IOTA’s real industrial use-cases. The focus is put onto machine-to-machine (m2m) interactions through the creation of proof-of-concepts (PoCs) for the emerging Industry 4.0. The vision is to enable “Smart Factories”, which are huge decentralized systems of independent machines that cooperate with each other in a “Manufacturing Economy”.

Using DLT for Digital Twins in the Supply Chain.

Choosing IOTA

Such a decentralized system requires an underlying data and value transfer protocol that is decentralized itself. Therefore, IOTA – as a light-weight, fee-less, permission-less, distributed, and scalable protocol for the IoT – fits perfectly into the entire picture. It is capable of secured data acquisition and real-time leasing of machine services in the Internet of Production (IoP). Further layers on top of the Tangle come in handy for exactly these purposes:

  • MAM – Masked Authenticated Messaging: a method that allows publishing of data on the Tangle in an encrypted fashion
  • Flash Channels: nearly instant micro-payment flows for real-time payments like pay-per-second
  • Data Marketplace: a decentralized marketplace that allows machines to buy and sell data
How MAM, Flash Channels and the Data Marketplace make IoP Applications feasible.

PoC: Digital Twins in Production

Products of industrial processes are usually not identical. Instead, they show slightly different properties due to production uncertainties and fluctuations in the material. With its first PoC, the IILA stores the individual properties (the digital twin) of each component on the Tangle. A web GUI can access the data for further inspection. Most importantly, this enables other machines involved in the supply chain to utilize this data, plan ahead for future production steps, and ultimately become more efficient.

Execution

To do this, the production machine owns a MAM channel. It publishes the datasets of the produced components in this channel. For this PoC specifically, the physical forces acting on the material during fineblanking (a metal cutting process for mass production) are measured. In a second step, the software extracts useful component characteristics such as the maximum force. The datasets then move to a queue which publishes them to the Tangle. Once published, the data can finally be read by the client directly from the Tangle and permanently stored in a local database. DynamoDB was used for the database implementation.

Scalability

The time required to do the proof-of-work (PoW) might become a limiting factor clogging up the queue, so the IILA decided to outsource the PoW to a remote server. They were able to reach 10 TPS during this PoC, although the system should scale even further (due to the PoW implementation via AWS Lambda).

User Interface

The front-end web GUI ensures data integrity by validating the signature of the dataset in the local database. This signature must match the signature published to the Tangle. However, the validation is not done automatically, as it requires too much time. Instead, the user can initialize it on demand by pressing a button. The GUI further allows to filter components by certain criteria and is able to visualize the origin of production on a global world-map.

The Web GUI Frontend: Visualizing Production Data.

Prospects

Unfortunately, the component’s data might exceed sizes practical for full on-Tangle storage. As a solution, one could establish a direct m2m connection to an external data server which stores the entire dataset. This would happen in exchange for a service fee to compensate the cost of maintaining such a server. Once again, verification was put in place to ensure data integrity: the signatures on the server need to match the temper-proof ones stored on the Tangle.

Additionally, in a future release the dataset will contain more complex data such as the heat dissipation (measured via Finite Element Analysis) during production. This will certainly require the aforementioned data server outsourcing simply due to its extensive size. With the planned integration of micro-payments, initial scenarios simulating those of a real m2m economy are feasible, namely pay-per-production and paid third party services (see the data server example).

Team

The WZL of the RWTH Aachen developed this PoC together with grandcentrix. In total more than 20 people (five backend developers, three frontend developers, two supply chain engineers, and two full stack developers, among others) engaged in the project. A good mix of different professions such as computer science, machine learning, cognitive science, data engineering, mechanical engineering and business administration ensures that the project meets quality standards in various areas simultaneously.

The IILA team as of April 2018.

Grandcentrix: The grandcentrix GmbH is a successful player in the emerging IoT and smart technology sector. They cooperated in this PoC with the WZL of the RWTH Aachen to work on realizing the Internet-of-Production vision. Their team brings in real world experience and development strategies in regards to IoT projects, and goes a long way to bridge over to the IOTA Foundation through key players such as Ralf Rottmann, who is both Managing Partner at grandcentrix and Member of the Board of Directors at the IOTA Foundation.

Community Feedback

After asking on discord, I received overly positive impressions from community members:

Check out their status reports. Not only is the project really cool, but their reports are a good read, too.
One of Germanys best universities working on IOTA usecases underlines the potential of the technology. And obviously is a “hint” that IOTA is not a scam. Really liked this article though, understandable even if you are not a super techie: https://medium.com/industrial-iota-lab-aachen-wzl-of-rwth-aachen/manufacturing-economy-e541066889ee
I want to visit Aachen lab, my parents live 40 km from it. (Sternn)

Closing Words

The IOTA community highly appreciates the IILA for its work in the IOTA ecosystem. By successfully applying the Tangle technology to the IoP sector, the IILA proves out IOTA’s original vision for machine-to-machine interactions. These real-world use cases go far beyond the typical cryptocurrency realm and highlight the fact that IOTA is actually a data protocol, and therefore much more than just another cryptocurrency. This kind of academic and industrial involvement is exactly what drives adoption and helps the IOTA technology to develop into a global standard.

If you liked this summary, I highly recommend you give the original posts of the IILA a look. They contain more visual material (images and videos), explain everything in much greater detail and, overall, are very pleasant to read: https://medium.com/industrial-iota-lab-aachen-wzl-of-rwth-aachen. You can also find them on the IOTA Ecosystem page.

All Pictures from the IILA blog posts.

SHARE IT!

Written by: Micro Hash

6 comment(s)

What is the actor of Flash Channel on this project? We know MAM could exchange data transfer between seller and buyer, but what’s the difference on pay with send_transfer API and Flash Channel? My guessing is : Send data with MAM and sign the payment txn at the same time. Finally, merge all the transactions into one Flash channel and PoW one time to attach to Tangle?

Flash Channels can be used for micro-payments when paying for a service such as the mentioned data service. If you send a normal IOTA transaction, you will have to perform proof-of-work and wait until confirmation. Both are expected to become negligible small in the IoT future. However, if you repeatedly use a service and want to pay-per-use or pay-per-millisecond, this would still be too much over-head. Flash Channels work off-chain and allow you to send instant micro-payment flows super efficiently.

Thanks a lot. There are three actors in your article (MAM, Flash Channels, Data Marketplace) and they are all decentralized. But in IOTA Flash Channel, you should put the signed transactions in off-Tangle before attached to Tangle. Therefore Is it necessary to deploy a cloud host?

Leave a Reply