Is IoT the game changer people believe it to be?
The internet of things (IoT) is set to become a $20-$50bn industry over the next 5-10 years. But what will happen to the incumbent industries that were remotely monitoring systems and controlling complex assets with computers long before 'IoT' existed? A lot of what we are seeing is nothing new, except the disruption caused by the end applications. In the past information was held locally in silos, but cloud computing and advanced analytics now mean that the silos are broken down and all data is available to generate insights. To provide some context, let’s take a look back over the last 60 years and see what else has changed and if we really have been here before.
M2M – back to the beginning
Arguably, machine to machine (M2M) communication started in the 1950s in military projects, where IBM developed the real time operating system concepts required to handle a number of complex applications, including the first missile tracking systems. In the same decade, industrial equipment was controlled, not by computers, but by hard wired relay logic. In 1968 a revolution occurred with the creation of the first programmable logic controller (PLC). Over the next 25 years this evolved into SCADA (supervisory control and data acquisition).
In parallel, developments were occurring to remotely monitor and control systems, mainly led by the utilities industry. In the 1950s remote assets could be controlled from signaling along fixed lines, and readings sent back along other lines. Cost per asset was high, but it did allow central monitoring, albeit needing human intervention at every step. In the mid 1960s minicomputers automated some of the processes which moved a step towards the central operation centres we see today.
Moving into the 90s
Another revolution came in the 90s with the introduction of Microsoft windows 3.0 and the accompanying Intel processors to support it. Suddenly there was a low cost solution to control problems that had previously required a very expensive minicomputer. The OLE (Object Linking and Embedding) software that Microsoft developed was exactly what was needed in both the SCADA and telemetry fields. In 1996 the OPC (OLE for Process Control) specification was released, which enabled a revolution in PC controlled systems in factories, and kick-started a revolution in Computer Numerical Control (CNC). Over ten years later, in 2008, a common framework became reality with OPC-UA (Universal Architecture) This has been used successfully by a number of companies to develop a rich ecosystem of compatible and interconnected products. An example is the water and oil and gas sector where millions of readings are collected every day.
Each of these steps was disruptive in their own way. Telemetry removed the need for remote teams at every site and each step of SCADA evolution meant that a factory could be run with fewer people. But compared to the disruption IoT promises these seem small.
Today – and why SCADA isn’t leading the IoT revolution.
Back to the present day, the landscape is littered with 50 years of proprietary protocols and vendor lock-ins. The SCADA world is now open and new companies have sprung up processing these newly available data feeds. Data is now sent over IP as well as old fashioned serial interfaces. Modern SCADA meets the definition of the internet of things in that you have many simple sensors connected over IP, so why is it not seen as the ‘gold standard’?
The answer to this question is complex. Firstly it comes from the assumption that everything about IoT is 'new' so has to be developed or redeveloped from scratch. Something that was put into standards eight years ago is clearly not leading edge and not applicable to the brave new world.
Secondly, markets are unlikely to change until a disruption appears to force them to. The incumbents in a market will have their cash cow products and services they want to maximize revenue from. As a result you will not see companies currently selling SCADA trying to compete with smart home heating companies such as Hive or Nest in the home space. There is no pressure yet to disrupt.
Thirdly there is a Total Cost of Ownership (TCO) question. If you look at the new technologies being announced almost every day, it seems as if the cost of IoT is falling, but in fact a number of the key Capex and Opex costs have remained the same for the last 10 years or longer. In particular the cost of installation and commissioning has not dropped appreciably. The incumbents know this, but the new entrants don't, until they discover that their TCO is not that different.
And fourthly, embracing and achieving real benefits from the new, analytics-led world requires true business transformation. Silos are jealously guarded as they are the source of job security and knowledge. If everyone has access to the data then the value of the data holder is reduced. In industries that have been using SCADA and remote telemetry for 40+ years, the guardians of the silos are entrenched.
But within this there is still the open OPC-UA standard, which is stable, proven, cross-platform and allows for local device discovery as well as cloud connectivity. We interact with it possibly on a daily basis, from the escalators in a railway station to the petrol pumps we fill our cars with. It would be the ideal basis for a home automation system or any number of IoT projects. Sadly it’s probably too open. In the gold rush of IoT, companies want new silos, IP and 'stickiness' to grow revenues and drive profits, not open standards.
Which leads us back to the beginning. The technology to collect readings from ’things’ has existed for years, but the holistic analytics were missing. In the old world a company having two factories would probably have two completely different systems running them. Insights were likely to be in the form of monthly reports. In the IoT world there will be the ability to run analytics to compare and contrast different factories. Secondly IoT is about the cloud, private or public. The level of interaction that IoT brings can only be supported with cloud thinking. And thirdly, change will be a much faster process. IoT embraces change. We have to just hope that lessons are learnt from the last 50 years.
Share your experiences and views on the evolution of SCADA, M2M and IoT with us…and what challenges and benefits IoT could bring to your industry.