Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Where Is IoT Headed?

DZone's Guide to

Where Is IoT Headed?

As far as IoT is concerned, the future is AI, ML, DL, predictive analytics with edge computing, and cross-sensor AI/ML models.

· IoT Zone ·
Free Resource

To gather insights on the current and future state of IoT, we talked to 23 executives involved with IoT. We asked them, "Where do you think the biggest opportunities are in the continued evolution of IoT?"

AI/ML

  • AI/ML learning from all of the data that’s being collected.
  • The value AI/DL (deep learning) brings to IoT. Smart sensors in buildings and trend analysis take effect and improve further by finding more patterns than humans can. What kind of DL patterns can we get? What model framework? Autonomous cars are route optimized.
  • At its core, AI is about simulating intelligent behavior in machines of all kinds—and since IoT is about connecting those machines, there is a clear intersection between AI and IoT. The benefits for businesses will be transformational. Given the power and scalability of AI solutions, tasks that used to take humans weeks or months to complete will be actionable in minutes or seconds.
  • 1) Continue to work on pilots and use cases to achieve business results. Service is moving to predictive on-demand work. Look more deeply to minimize downtime and service costs. AI/ML/predictive optimizes uptime while reducing costs. Long-term there’s a foundation for transferring to more holistic business models where you are connected to the supply chain to meet demand in real time. Custom products will be created and shipped with autonomous vehicles. Energy, infrastructure, large manufacturing will be using microgrids for generation and distribution. This will impact how we plan for improving infrastructure and having a smarter infrastructure.
  • Edge computing with AI/ML and predictive analytics.
  • The biggest opportunities are cross-sensor AI/ML models, tapping into the yet untapped capabilities of existing systems, and the use of more untraditional sensors like till systems.
  • Making user engagement as direct and valuable as possible. More automation on the edge to leverage data with predictive models that create better value for the customer. You know what the customer needs before the customer realizes it and you just provide it – anticipating and meeting needs.
  • What happens a moment ago impacts the enterprise a moment later. Continuously learning how to be proactive and change how business is done. 1) Ingest everything in real-time. 2) ML models run more frequently. 3) Business can react more quickly. Ingest, analyze, learn, and deploy in an operation application seamlessly and fast so the company can change proactively.
  • Local application of global communications. Higher level predictive analytics. The distributed model needs to be deployed in industrial. PLC (program logic computer) brings a more flexible platform for additional analytics. Subscription level services to access data anywhere in the world. Data in motion from remote areas.

Industrial Applications

  • While there may be more share of mind in the consumer sector, there are greater opportunities for revenue in industrial/enterprise – factory floor, remote monitoring
  • The greatest opportunities lie in software companies, hosting companies, and non-traditional software players partnering to develop OEM solutions. Software companies participate in industry vertical solutions to handle scale and software IoT specialties and nuances.
  • So many applications, I’m not sure what will be successful. The market will determine this. Do people really need/want home automation? I see more opportunity in industrial automation – factory lines, smart storage, power grids. We have a shipping client that is building a ship with all connected parts so the ship can be controlled remotely. Factory lines will be streamlined. 
  • The industrial infrastructure to entire cities, fields, and regions can be monitored. We’re at the early stage of build out.

Other

  • Meet people where they are without network or business boundaries.
  • Accelerating growth. Ubiquitous access via the cloud.
  • Innovations in battery technology so you reduce the frequency of recharging. MCU providers with low currents and voltages.
  • New experience and business model. Rely on IT companies to ensure the system works, is secure, and able to manage lifecycles from five to 15 years. Customers are confused by the complexity of it all, including the providers. We will see consolidation by the best of breed solution providers that clients can trust.
  • Integration across devices whereby there is a standard identifier for each individual so that each individual sees real value from the IoT devices with which they interact.
  • As the IoT revolution continues to gain momentum and each year sees exponential device growth, the opportunities are seemingly endless. From a consumer perspective, IoT development should continue to fuel automation that furthers everyday life conveniences – from smarter refrigerators that know when you’ve run out of certain products, to automobiles that sync to online calendars to know exactly which appointments they should be driving to. From an industrial perspective, connected sensor technology should continue to be developed and deployed in items such as electricity meters, traffic lights, street lights, etc., in order to lead society into a smarter and more sustainable way of operating. As IoT devices become more pervasive in everyday lives, having a sound approach to security becomes increasingly critical. However, for the public to take IoT security seriously, it may require something similar to what happened in the late 1990s when the world’s PCs began connecting in scale to the Internet due to the introduction of low-cost broadband connections. This, in turn, led to the widespread proliferation of viruses and malware, which gave rise to a completely new industry of AV. But, the AV market didn’t truly develop until there were a series of headline-grabbing virus attacks that created havoc. That same scenario may well play out today, but with the difference that today’s attackers have the owners of billions of web-connected devices at their disposal, combined with advanced AI to generate attacks that the hackers of the 1990s couldn’t begin to imagine. As it did in the late ‘90s, there will be considerable opportunities not just for the hackers, but for the device manufacturers and security vendors who are able to best and most effectively respond to these threats and provide protection for their customers.
  • I'm concerned with connectivity. LARA, SigFox, cellular are getting better. We need communications protocol standards for easy, global connectivity.

Here’s who we spoke to:

Topics:
internet of things

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}