Like many people I know, I find the whole M2M/IoT space confusing. Even though I have worked on many M2M solution designs myself (car parking machine ticket paper and coin box monitoring, hire generator fuel levels, run time, ODB2 and environment monitoring, remote ATMs cash box monitoring and connectivity, and volcano environmental monitoring stations) using Peplink equipment.
Naturally, the underlying network technology is IP on cellular data networks (as that’s all we currently support). Alternatives to cellular operator networks are, of course, of interest to me.
Personally (and hopefully accurately) I distinguish M2M as the underlying connectivity method and messaging technology (the ‘I’ in ‘IoT’) and IoT as the ultimate result (and intelligent use) of the data collected from these devices (the ‘Things’ in IoT ) that is received from (and sent to) the devices using the M2M networks from web-based data applications and dashboards consumed by users.
So, three big pieces there: the ‘thing’ or product/device that does the data collection; the M2M part – how it connects to and via multiple networks to get that data somewhere to be analysed and actioned; and then the data manipulation part that turns the device’s data into useful, commercially valuable, user-actionable information that justifies the whole purpose (and cost) of the IoT deployment in the first place.
Taken individually, there are no insurmountable challenges facing the adoption of IoT from any of these parts. New approaches are appearing every day to make each element more efficient and capable, whether that be new lower-power edge devices and wireless communication technologies, scalable transactional messaging systems, time-based data storage solutions or front-end application frameworks.
From experience I know that when you put the network operator, hardware engineer, software developer and UX designer into the same room and ask them if any particular IoT application is technically feasible today they will all say – yes.
No, I think the biggest challenges facing today’s IoT projects is solution-based risk mitigation and ultimately proving and maintaining the complete solution’s commercial viability.
This is evident I think in the current approaches that everyone in the IoT space is taking to standardise as many elements as possible in an attempt to speed up time to market by reducing total project complexity.
Everyone is trying to build products and services that will plug and play nicely with others to speed up project development and deployment times – which also mitigates risk since any element from any provider could be subsequently swapped out for another compatible product or service if required for commercial or technology reasons.
I’m certain that no one really knows yet which standards and methodologies will win out in the end, but everyone knows that compatibility and interoperability at every level will be a key proposition of the most successful solutions.
There is a big opportunity here I think for hardware and software interopt vendors who can act as the glue between the currently available hardware and software platforms and abstract those elements to reduce the risks of technology ‘lock-ins’.
As an engineer though, my biggest struggle with IoT’s expected explosive cross vertical growth is actually understanding the commercial drivers for IoT projects to begin with. Of course there are numerous examples of obvious cost-based benefits to IoT solutions in narrow verticals.
Home electricity meter monitoring, for example, adds value to both customer and operator by enabling the home owner to manage their consumption and the operator to reduce their fleet of in-field meter reading staff – saving everyone money and paying for itself in short order.
However, when you move to the full paradigm of internet connected everything (the billions of devices that everyone screams at you are coming), including coffee machines, toasters and waste paper bins, the commercial drivers are much harder to grasp.
Internet-connected waste paper bins, for example, suggest a scenario where janitorial staff in a commercial building can focus their waste paper collection efforts on those specific bins that are full.
Try explaining that to a building’s janitor though and you can expect a shrug of indifference because he and his staff have to visit all areas of the building anyway.
Like most non-vertical specific IoT solutions, they only become commercially viable when deployed on a large scale and then subsequently targeted at a specific vertical.
Intelligent waste paper bin monitoring makes most commercial sense, when many hundreds if not thousands of bins are being monitored by a single specialist waste paper collection service that would find commercial value from not needing to visit each and every bin they monitor every day.
And this is ultimately where my confusion comes from. All of the big players in the M2M/IoT space are positioning their products and services to be suitable for their specific definition of IoT deployments across all verticals.
Whether their products and services are actually suitable or not depends on the customer’s definition of IoT in the end solution that they need to support, and to be commercially viable any end solution has to be based on the specific requirements (and commercial drivers) of a specific vertical – all of which are vastly different.
Arguably then, the biggest challenge still facing the wider adoption of IoT is the recognised definition of IoT itself, depending on where it is used and by whom.
At a business level, IoT solutions will only be adopted by enterprises when they add commercial value to their business proposition, and so for them IoT is always about specific business intelligence in a specific vertical that adds commercial value to their business activities – that justifies the cost of deployment.
For consumers and end users, however, the real value of IoT comes when data from individual business IoT solutions in specific verticals is aggregated to add value to (and simplify) their lives.
This is the big picture of IoT that we are all being sold now – when your smartphone tells you before you leave work that there is traffic on the motorway and so advises you take a different route, but knows that you don’t have enough fuel in your car so re-routes you via a service station, and knows from the scanner in your fridge that you are nearly out of milk in the office so you need to buy some when you fill up.
In my opinion, IoT adoption to the point where billions of everyday devices are all interconnected intelligently is a two-stage process. Businesses will first adopt the benefits of IoT to add commercial value to their activities in their specific verticals – creating pockets of really useful but highly focused data.
Then consumer IoT service providers will buy access to this business data and subsequently build and sell consumer-targeted IoT services that link these business data sources to provide intelligence that adds value to the consumer.
Explosive IoT growth is coming – of that I’m certain – but I think it will be far more like an extended fireworks display over the next few decades, than the single earth-shattering explosion we are all being told to anticipate in the next few years.
About the author: Martin Langmaid is a solution architect and technology evangelist at Peplink, focusing on cellular M2M connectivity solution design. He specialises in network design using fixed and cellular multi-WAN link aggregation and load balancing for resilient, secure diverse M2M and WAN network deployments.