Some mistakes to avoid in your next Industrial IoT project
August 25, 2020, by Peter Rawlins
A study revealed in 2018 that only 26% of respondents considered that their industrial IoT projects were successfully completed. Here, some potential reasons for failure.
There is a lot of talk about the importance of IoT as an exponent of current economic development, particularly for the evolution of industry. But sometimes, things look simpler from the outside than they really are. A Cisco study revealed in 2018 that only 26% of respondents considered that their industrial IoT projects were successfully completed.
There are many reasons why this success rate is so low, but what is clear is that most projects fail in their initial stage. This is what we, in Barbara IoT, have baptized as the “Proof of Concept (PoC) Purgatory”. We refer to that place where projects stay forever in the testing phase of a few devices, without ever becoming implemented on a large scale.
This is unfortunately common; industrial companies face multiple challenges during the start-up phase of a project, from the establishment of the right alliances up to the choice of the right technologies. However, the greatest difficulty will come later; morphing from a few prototypes working in a controlled domain of limited scope, to a multitude of dispersed devices in complex environments with varying scopes.
Let us take the example of IoT deployment for solar energy. During the “proof of concept” phase, the initiative may establish communications with a solar energy installation and move inverter and other performance data to a cloud. However, large-scale deployment covering numerous solar parks will almost certainly reveal a host of disparate technologies used in the installed base. These technologies will be using distinct legacy protocols with implementations differing according to use. Solar farms will have communication snags, and in some the radio spectrum may be convoluted. Situations get ever more complex as large-scale projects roll out.
This array of issues is demanding to foresee and to validate during the PoC phase. And believe us when we say that one single technology will not be able to address the real-world complexities unless we start addressing these concerns from the beginning. In Barbara, we think there is a way to do it.
We highlight three main themes that a business need bear in mind when tackling an Industrial IoT project with a view to avoiding future failures:
● Device Fragmentation: Plan to be able to support sensors and protocols of different suppliers. Do not expect a single manufacturer or technology to cover all your use cases over a lifetime, even in any one installation. To do this, design a layered architecture of independent components for the PoC. A three-layer design of “things-comms-cloud” will provide the necessary modularity to integrate new use cases along the way.
● Vendor Lock-in: using a single vendor that provides a monolithic and vertical approach, providing each and every piece of the puzzle, is likely to lead to failure. No one can be master of all trades, and even if they could, do you really want to be at their mercy? It is therefore pivotal to incorporate solutions based on open standards and interoperability at the PoC stage.
● Security: last year saw a 300% surge of cyberattacks on IoT devices. Edge devices are the weakest and most attractive entry point for cybercriminals. Instead of using insecure by nature devices dictated by pricing considerations, secure by design IoT gateways should be imbued over all the network to counter this threat. Taking security seriously during the PoC phase is a project life insurance.
At Barbara we believe that there is a different and better way of doing industrial IoT. The PoC is not the end of the road, but the beginning of the value chain.
Suscribe to our news letter to stay on top of the Industrial news and IoT trends