(86) “How Data Moves in a Modern Factory”​ by Kudzai Manditereza, Product Architect and CEO, Cloresol

I just finished recording an insightful podcast episode with Kudzai Manditereza, Product Architect and CEO of Cloresol, an IIoT based company in Johannesburg, South Africa. Kudzai writes a lot on technology subjects particularly Industrial IoT and Industry 4.0 and that too in a very interesting manner. And that is how I got to know of him — through his writings. So he began exploring the implementation of IIoT solutions and In 2015 he founded Cloresol, which he currently run as the CEO. And he has been focused on making Industrial IoT available to SME manufacturers and in that regard we’ve been Integrating various IIoT Systems for our clients and also providing OEM manufacturers with products that make their machines Industry 4.0 compatible.

The topic of our conversation was, How Data Moves in a Modern Factory.

Kudzai has shared an infographic on the same topic on his Linkedin blog which had caught my attention. I approached him and asked if we could make that as a topic of our conversation.

Here is Kudzai’s infographic:

Here are the key insights of our conversation:

• What is Cloresol’s business focus?

Cloresol’s business is focused towards digitally transforming small to medium manufacturing companies. That’s because Industrial IoT is often spoken about in the realms of large organizations but we want to shatter that myth.

We also building Industry 4.0 compatible products for OEMs because most OEMs use legacy systems that are not IIoT ready.

• Help us understand the logical divisions of these three networks — Plant network, Operations network and Enterprise network?

Connectivity at the factory floor can be divided logically into three layers — connectivity at the end (plant network), operations connectivity and connectivity at the cloud (enterprise).

With edge or plant network, we are dealing with physical devices and components like motors and PLCs, Gateways, assembly line etc.

Operation network deals with various software that work in conjunction with the various devices in the factory floor.

• What does a typical factory floor look like?

The equipment in a factory floor can be classified into different types depending on type of connectivity. Two wire protocols(I2C) machines, which could range from pumps, valves, blowers or temperature sensors. There could also be machines such as compressors, robotic arms, and various PLCs from different vendors.

In most cases, these PLCs are not designed to communicate with each other, especially when they are from different vendors. So most of these devices were not designed to work on Internet and were built for constrained environments.

There are other class of devices that were designed to work with open technologies and standards like OPC-UA. These devices can communicate with other devices in a plant.

• How do you communicate with devices that have proprietary protocols?

It is a challenge to communicate with such devices and to get them to communicate, you need protocol drivers written for such devices on the Gateway that converts all of that information into a standard communication protocol like MQTT or OPC-UA.

• Where is the next level of innovation going to come in the IIoT architectural stack in the future?

I have written a blog post on that already — https://www.linkedin.com/pulse/why-iiot-revolution-happen-edge-cloud-kudzai-manditereza/. The next innovation is going to be at the Edge. Most processing will move away from the Cloud to the Edge. For argument’s sake what we perhaps need to establish before moving on is where or what is the edge?In short, the edge could be a sensor, pump, car, gateway. Any point where compute and storage could exist.

Sending terabytes of data to a central point for analysis and decision making and then sending it back down again will never work. It is not sustainable both in terms of bandwidth cost and increase in network latency. Time to insight and action can be tremendously accelerated only if machine learning models, server-less computing and lightweight databases are driven into edge devices.

Sending terabytes of data to a central point for analysis and decision making and then sending it back down again will never work.