Utilising IOT in your business can be one of the most significant levers to pull to achieve measurable value like cost reductions, optimisation or delivering specific outcomes like automated compliance or worker safety.
The definition of IOT can be broad but, put simply, it is many small devices sending tiny payloads of data, all working cohesively, to create actionable data and a solution that allows for remote access and sometimes remote manipulation. An example use case might be using IOT sensors to monitor soil moisture and a rain gauge to trigger irrigation when necessary.
There are multiple applications for IOT, and due to broader adoption, the cost of implementing an IOT project has come down in recent years. Logically, one assumes that with more companies using IOT that the complexity will reduce. In my experience, however, the process of delivering a successful IOT project can still be very complicated. Implementers of IOT often underestimate the scope of work and are surprised by unexpected complications.
IOT is a fantastic field where the growth being experienced is phenomenal. Still, it’s not necessarily the right option for every use case. Encouragingly, the number of devices is growing exponentially and there is an excellent quality sensor for every type of measurement. Some of the most successful projects I have worked on use sensors that are purpose-built for one or two specific measurements, with multiple sensors working together as a whole to make lives easier and businesses more profitable.
Hidden challenges
While IOT’s potential is huge, there are still challenges. Actual implementation and project roll-out can be slowed down significantly by unforeseen complications – usually due to lack of understanding around how the various systems fit together to create a viable solution.
IOT products don’t necessarily follow the same standards, even if they use the same protocols – for instance, LoRa. This is because every product manufacturer develops their product in isolation. To make matters worse, some IOT products are actually rehashed outdated products equipped with LoRa communications chips in the hope that they will be more successful in an IOT era. This means that the products are already old technology and not built with the same principles in mind as IOT predicates.
Therefore, integration can be the biggest challenge: projects that were scoped for weeks of development suddenly take months of preparation just to get to the point where they can be integrated. Something as simple as a LoRa sensor to send temperatures to a server can involve far more integration complexity than expected.
The LoRa devices don't send data directly to a server for consumption. They have to go through an LNS (LoRa Network Server) such as TTN (The Things Network) or Helium and then you first have to understand how each different LNS works in order to figure out which best serves your purpose. The organisation could use Chirpstack, Loriot or WMC (Waney Management Center), but all of these options have costs involved. In the grand scheme of a good sized project, they may seem insignificant; however, they are recurring costs that should not be overlooked.
Hidden costs
Almost every system that sends data to various disparate systems will have been built at a considerable cost and that cost needs to be recouped. The amount of data being sent and stored also needs to be accounted for, meaning more costs. Visualisation of that data has to be realised through even a different system, with yet more costs.
Delivering value
In light of the unexpected challenges and costs, one of the greatest mistakes a company can make is to try and implement IOT just because it's such an industry buzzword.
Whether IOT actually adds value to a company should be the one and only driving factor to implement it. Whether it adds value depends on the use case.
For example, using an IOT solution to monitor a building’s electricity consumption remotely will most likely deliver little in the way of value and savings. However, using the same electricity meter to bill tenants more accurately will save money and deliver value. A use case that could really add to your bottom line would be to distinguish between grid power being supplied to your tenants versus the more expensive diesel generator power provided during load-shedding, and billing tenants accordingly.
Another use case that makes real sense and adds tremendous value is predictive maintenance on regularly serviced machines. A sensor as simple as a LoRa-based accelerometer can use AI to predict when a moving part could be failing. The typical AI-based sensor will run for a short period of time, learning what the normal vibrations mean on a machine; when a bearing is damaged, even the slightest vibration out of kilter can be monitored and an automatic service request can be actioned or some alert can be sent to a maintenance technician to turn the machine off to ensure further damage doesn't occur.
Importantly, the correct parameters need to be measured to ensure the desired outcomes.
IOT not a catch-all
IOT is not a catch-all for every use case or situation. The use case needs to be carefully considered and the outcome should be that it saves money, travelling time or some other costly parameter.
However, this doesn't mean IOT doesn't deliver value. The technology has proven results in thousands of cases around the world and it is becoming more evident that when applied intelligently, IOT provides immense value.
We're seeing more and more standards that are being introduced, and IOT is maturing, with significant advances in IOT products. I view this as a development that could make IOT more viable for more use cases. I believe this will bring stability to what has been a 'wild west' IOT landscape.
Share