Some of the more optimistic futurists have predicted that in the not-so-distant future, the world would be amazing and beautiful; where our ancient ancestors would have thought that that world would be paradise. Depending on your level of optimism, we may actually not be too far from that paradise.
Humankind has experienced unprecedented advancements in technology in the last 50 years fueled primarily by the Internet. Technologies like Artificial Intelligence (AI), Machine Learning, and a flourishing number of gadgets being accessible to answer our every whim and fancy, is helping to shape that utopian future.
To give some perspective on the Internet of Things (IoT) being the “next big thing”, here are some predictions for its future:
- Bain predicts that by 2020, annual revenues could exceed $470B for IoT vendors sellinghardware, software and comprehensive solutions
- McKinsey estimates the total IoT market size in 2015 was up to $900M, growing to $3.7B in 2020 and has a potential economic impact of $2.7T to $6.2T until 2025
- General Electric predicts investment in the Industrial Internet of Things (IIoT) is expected to reach $60T in the next 15 years
- IHS forecasts that the IoT market will grow from an installed base of 15.4 billion devices in 2015 to 30.7 billion devices in 2020 and 75.4 billion by 2025
So what helps us accept, adopt and embrace IoT? What is the impetus for this industry? What spurs us to wear a smart watch or fitness tracker or even install a smart light at home?
The rise of IoT is not a fad; there is a growing need for autonomous control that can be soothed by having the right IoT deployment solutions. Through this control, there can be added efficiency, with data collected from IoT processes used to make better and more informed decisions.
We are always in constant pursuit of happiness. It becomes basic economics to maximize the utility function, and IoT is able to help us achieve that, and this is probably the reason why we see such a diverse range of applications and the huge potential in IoT.
Imagine sensors being able to tell the individual energy levels of someone at work, school or home, with personalized preferences that detect when the person is starting to feel tired, after which the trigger is sent to the massage chair to start up. Maybe the coffee maker could even start brewing a fresh cuppa on its own, or a subtle non-lethal wake-up shock is sent during a boring lecture to wake that person up: this is the almost-perfect world of IoT.
However, time for a reality check: we are still not quite there yet. Unfortunately, good IoT solutions and applications are oxymoronic – simplicity is not without its complications.
At the first connectivity layer, if we pick to use wireless, depending on the range, there are several technical standards to manage and master.
Complexities of IoT
Currently, IoT practitioners like Overdrive need to navigate through a few levels or layers in order to implement a device for a proper use case. The landscape of IoT from device or sensor to the final user application typically starts with choosing the type of connectivity, then link and transport protocols, session and communication handling, data aggregation and basic processing such as filtering, then data storage and retrieval, which needs to fit sensibly into the business model, before finally being extracted into the applications.
This multi-layered complexity may in time be flattened by a superior, leading platform, which Overdrive aspires to be, that helps to manage several of such technicalities through a single platform. But until then, this may be one of the main reasons why even though there is a sudden explosion of consumer applications, a true ecosystem of devices still does not exist. To this day, most devices include a limited range of sensors and are only suited for a singular application instead of being able to work and “talk” with the other devices around it.
Meaning: this is why even though your smart watch is the pinnacle of a smart device, it is still not able to communicate and synergise with your washing machine, refrigerator, air con, lighting system, TV or your car.
The current IoT landscape is akin to the United Nations (UN), except it can be more complicated for the fact that the UN consists of a few hundred member-states whereas
the IoT world may have a few thousand different types of devices and growing! All the sensors and device are like people with different cultures trying to work together while maintaining their own identity, or their own purpose.
While the current IoT landscape is rather complicated, nevertheless, the IoT ecosystem is undoubtedly thriving and alive with tremendous opportunities.
The Gap From Now Till IoT Paradise
There are several challenges that practitioners like us, Overdrive, still face in deploying meaningful IoT applications.
Not all devices are made equal
Everyone assumes devices are smart. There is no doubt that there are tremendous technology advancements in the design and development of the microprocessor chipset, and the world is looking forward to the advent of quantum computing. Currently, many devices such as the smart phone, are operating on a 32-bit OS (Operating System).
Nevertheless, there are simple but still useful devices operating on 8-bit OS, as well as some simple single-purpose sensors. There are cost and application factors where it may be more efficient to use such simple devices and sensors. The entire IoT landscape has to be flexible and encompassing, to allow the range of devices and sensors to work together. This is important as it will average out the workload and lower unnecessary overheads in the entire ecosystem.
Standards, standards, standards
Don’t be surprised to find that there are proportionately as many government bodies, conglomerates, cooperatives, associations and academia communities working on defining the standards for the use of IoT as compared to the number of devices in the IoT universe. While standards allow a greater degree of interoperability between IoT devices and applications, we are still at the early stages of discovering the full potential of IoT, therefore the standards should not be restricting innovation and growth, or worse, defining something which may turn obsolete.
IoT is seeing growth primarily because of the “Internet”, quite literally. It is interesting to find the appropriate sensor or device to capture the signal or readings from a certain machinery or micro-environment, turn it into a bunch of data, but in many real world situations, the challenge is in connecting this data with the world. Let’s illustrate with two examples: we know that IoT can play a major role in transforming traditional agriculture practices by introducing sensors to monitor parameters such as soil nutrient, sunlight, water levels and so on.
But many existing traditional farmlands are well out of reach in rural areas with limited connectivity. It would take additional resources to deploy suitable infrastructure to collect the readings from sensors deployed all over the farm.
Another example: IoT can also make a huge impact in enhancing the quality of service that can be provided in aged care. There are wearables with suitable sensors that can monitor heart rate, blood pressure, movement and activity readings. However, many of such wearables have limited connectivity options to send all the data to a central repository where better services can be deployed to render assistance to the wearer when in need.
Security and Data Privacy
With a growing number of devices connected to the internet, questions arise on data security and privacy, which has to be addressed by more than just disclaimers and a barrage of ads. Who do we allow to know what needs replenishing in our smart, connected refrigerator when paired with a major e-commerce grocery site? Who needs to know how many steps we have taken or how many cups of coffee we have consumed? It is a balance between convenience and protection against malicious intent.
Sharing Our Experiences
Through the deployment of The School Bus Project, here are the takeaways we would like to share:
IoT is like Lego blocks
We liken the deployment of an IoT solution to building with Lego blocks. There is a need to balance the technical feasibility with stakeholders buying in. Start with a simple but workable solution, then keep improving or adding on modules. We started with a contact-based School Bus System using simple RFID modules where each student needs to tap on when boarding or alighting from the bus. Although it was only one of the ways this could be done, we were still able to provide some data visualization relating to the ridership by generating trip reports.
The next enhancement was to use Bluetooth iBeacons which made the entire process a contactless journey. Students carrying our Bluetooth iBeacons simply just boarded or alighted, and through our Bluetooth base stations on the bus, the system was able to tell who was now on-board or has alighted. We were able to discover and successfully deploy Bluetooth technology as we focused on making small yet significant wins through first implementing a simple solution.
There was a gap of roughly a year when we first deployed the School Bus Project using RFID before we recommended and deployed the enhanced version using Bluetooth technology. This is because we initiated that our users had to digitize and automate some of their current systems and processes so that their operations could better adapt and handle the new streams of IoT data that come in.
Without this strong foundation, the data acquired through the deployment of IoT may not reap the benefits and may even be an extra burden. While IoT is not a miracle cure, it can be very powerful when utilised properly.
IoT May Be The First Touchpoint
In many ecosystems, IoT is probably the first touchpoint, the “source” of the data. Therefore, no effort should be spared to really understand and investigate as much as possible the operating environment, and the suitability of the type of devices or sensors to be used in that environment.
As IoT may be new to many, we highly encourage users to explore, be open, and even take a proactive approach in experimenting various combinations and options. Taking the School Bus Project as an example, we evaluated the pros and cons of other options such as active RFID, camera systems, smart phones and even customised smart wearables before finally deciding on Bluetooth.
Good IoT, Good AI
To achieve advanced business insights, quality IoT data input will lead to quality analysis. IoT and AI are not conflicting or competing technologies, but in fact complimentary. For example: a smart watch does not automatically increase your fitness, it is the decisions that you make with the data. So in certain situations, there is a need to identify and decide which is the horse and which is the carriage.
Conclusion or Inconclusion?
IoT is still in the early stages of realising its full potential.
If anyone should try to peer into a crystal ball to see what excitement there is to look forward to in the development and evolution of IoT, the future might be shrouded in a cloud of uncertainty.
Right now, IoT does not seem to garner as much interest compared to the other flavours of the week such as AI, machine learning, or drones and robotics.
Furthermore, comments made by prominent figures and business leaders, whether sensible or otherwise, add to the unnecessary stress and noise that hinder us from properly understanding which area or industry each field of technology is most suited for and can perform best in. Something else to figure out includes the entirety of the technology landscape, more importantly, how such technologies can help us and what it really means to us.
We are currently being blasted with the image of how machine intelligence will soon surpass humans, just because machines are able to beat us in a game. This, for example, becomes a distraction which does not do any justice to the further proper exploration of the technology but instead just hypes up anticipation and fanfare that AI and machine learning is seemingly able to perform miracles.
Without properly referring and giving due consideration to fundamental basic building blocks of technology such as IoT, this will affect the practical application of AI or ML in any situation.
Overdrive takes on the approach to be innovative and pragmatic. If we are to reimagine what we foresee the technology point of singularity would look like, we see sensors and devices communicating and interacting between themselves, sensors triggering responses on other devices, where machines truly learn, adapt and respond to what is happening in their operating environment.
The different fields of technology, whether IoT, AI, or quantum computing, will need to adapt and grow not independently in isolation, but evolve together, perhaps with a substantial degree of cross pollination amongst the different fields.
We look forward to a time in the future, whether AI or ML, IoT or drones and robotics, are not different factions competing against each other; where names are just names.