Recently, analyst firms Forrester and Gartner released their round-ups of the most important technologies to watch for in 2017 and the following years. Many of their findings resonate with the manufacturers and other companies we meet every day. If we were to offer our take on the trending innovations you should think about as you set your technology strategy, we would pick five closely related technologies that have one thing in common: They help us decide and act strategically and wisely by collecting, processing, and analyzing large amounts of data.
1. Internet of things (IoT)
2. Machine learning and artificial intelligence (AI)
3. Intelligent agents
4. Intelligent things
5. Augmented reality (AR) and virtual reality (VR)
Here’s how we see these technologies make their impact.
From sensors to insights
Imagine connecting the devices and machines in your operation, as well as complex products you deliver or install for customers, to the IoT or the industrial internet of things (IIoT). Data from the sensors on these industrial assets is collected and stored. Because you need to make sense of the information from your devices, machines, and products, the IoT and IIoT are inseparable from the analytical and data management software tools that let you make sense of the big data you generate. Such solutions include innius, our IoT-based machine intelligence solution.
What happens when you apply analytics to IoT data depends very much on your goals and interests. You can turn sensor data into a real-time insight based on a well-weighed review of all sources and the data patterns that become visible. The findings presented by your data analysis solution can help you improve the throughput, performance, uptime and productivity of your machine assets. What’s more, you may be able to use them to improve the quality of the output, or close the loop between your installed products on customer sites and engineering by improving your designs or developing valuable innovations.
To simplify your data analysis, innius, our IoT-based machine intelligence software, includes several standardized KPIs, including Overall Equipment Effectiveness (OEE). Whenever you run the metric, innius gives you an up-to-the-moment view of the OEE of your machinery, based on the data streams from many sensor sources.
Evolving toward AI
Machine learning can consider the data streams from your IoT-connected machines, combine them with information from your ERP, PLM, CRM, and other business systems, apply advanced algorithms and analytics, and present useful insight. Could you accomplish the same with human intelligence? Maybe, but it would take much longer, introduce needless errors in weighing data contextually and accurately, and our tendency to judge based on our flawed intuition would also get in the way. A popular usage example for machine learning is predictive maintenance: Based on findings from IoT data analysis, your software tools draw your attention to a certain condition—think of equipment vibration, or wear and tear on a critical machine part—that indicates a potential problem, before it disrupts your production. You can then initiate the right repair or maintenance action.
Going a step further, AI could learn from the issues, analyses, and predictions of machine learning to act on your behalf. At that stage, you don’t just keep your machines in great running order, you improve your operation. For example, AI might see a way to optimize your OEE by making an adjustment in the production rhythm or the dimensions of a product part.
Agency comes to industrial assets
Once you gain the IoT-based, data-driven intelligence from your analytics and delve into the basics of machine learning or even AI, intelligent agents are not far off. These interactive solutions would access the data insight and machine learning related to your connected industrial assets, and provide answers in response to your questions. Instead of writing a query, you can simply use natural language. For example, imagine you have a connected machine placed on a customer site and innius indicates that the output is approaching the extreme ranges of acceptable dimensions. If the trend continues, you face quality concerns that may require shutting production down. You ask your intelligent agent when a technician with the right skills is available and close to the customer location. After answering your question, the solution can also schedule the technician for onsite visit at the next opportunity. If the technician can rely on an intelligent agent, the agent could advise her on best maintenance or remedial measures, based on the collective experience of your engineers and maintenance teams.
Over the next few years, ever more effective and ubiquitous intelligent agents will have a great impact on our workplaces and communications. Such solutions as Microsoft’s Cortana or Siri from Apple are evolving quickly. Today’s chatbots, virtual personal assistants (VPA), and virtual customer assistants (VCA) will take an increasingly proactive role in anticipating people’s needs and decisions.
What if an intelligent agent is embedded into one of your machines? With intelligent agent-capabilities, powerful IoT data analytics, and machine learning, you connected industrial assets can become intelligent things that can interact with you. They could assist issue resolution and provide guidance in such areas as the utilization and capacity of your machine lines, your materials planning and procurement, or the compliance of your food manufacturing plant with allergen management policies. Intelligent things will require sophisticated analytics and advanced machine learning to make sense and have the right focus and purpose. In this context, deep learning—a quantum leap in the power of machine learning—will become more practical and versatile for industrial and enterprise applications. Deep learning applies sophisticated algorithms to extremely large masses of data in digital neural networks.
The digital empowerment layer of AR and VR
IoT data insight and machine learning come full circle in a digital overlay of AR or VR over the physical world before you. Your intelligent agents and intelligent things can benefit from AR or VR built for them as much as you do. In an AR scenario, you can turn toward your machine, and your mobile or wearable device tells you how well it performs, what its recent output and quality is, and what maintenance it needs. It can then guide you in using the right tools to access and replace the appropriate parts, enabling you to do so without taking time away for training. Or, AR can help you find out when you can expect the warehouse item on the shelf in front of you to be out of stock, what your options are for reordering it, and which vendor can offer the best quality for the most reasonable price.
In product engineering, VR is rapidly becoming an efficient, economical collaborative resource that allows contributors anywhere in the world to design and enhance such complex products as vehicles, planes, or robotics. In the consumer realm, shopping experiences and decisions will become enriched with AR and VR to inform choices and let people try out products virtually. Behind the scenes, more and more powerful analytics, including spatial and environmental analytics, ensure that AR and VR are agile, practical, and relevant.
More companies, including smaller organizations, begin using these technologies starting with the IoT, where the entry threshold is low. With innius, for example, you can begin by connecting a single machine, adding other assets by and by as you see how the machine intelligence brings value to your business. As companies move along the road to digital transformation, they will find their own, unique mesh of the technologies we discussed.
Do you agree that our top five offer the most promise of improving and transforming our businesses and workplaces? At To-Increase, we are planning to take full advantage of them in helping manufacturers and other businesses undergo digital transformation and maintain their viability and competitive standing into the future.