IoT in manufacturing: The four stages of maturity
- 01 April, 2017 06:25
Although vendor-written, this contributed piece does not promote a product or service and has been edited and approved by Network World editors.
Only a couple of years ago, IoT was still shrouded with intrigue and uncertainty. The integration of sensors, connectivity, and big data technologies made sense in the business world, but their true potential, implementation, and use cases were far from clear. There was a lot of talk about when exactly IoT would “hit” businesses. Today, instead of having a single demarcation point for IoT’s acceptance in industry, we’re seeing manufacturers adopt it in chunks. Quietly but steadily, IoT is reshaping different parts of the manufacturing process.
The use cases in industries like aeronautics and chemicals are a proving ground for the real-world potential of IoT. There’s a roadmap to IoT adoption beginning to form. It often starts with Enterprise Asset Management and goes from there, garnering more revolutionary potential along the lines of visibility and automation.
While individual companies will have different approaches to experimenting with and deploying IoT, we can break down the journey towards IoT maturity into roughly four stages:
1. Enterprise Asset Management (EAM): EAM is an approach to managing industrial assets holistically through the use of software. IoT has already come into play with regards to “asset performance management,” using IoT sensors and connectivity to understand and predict when equipment will need maintenance or risk breaking down.
By equipping industrial machines with IoT technology, companies can access waves of real-time data regarding performance, workload, stress, and a host of other significant variables. Analyzing this data, it’s possible to correlate factors that lead to equipment failure (including external factors like weather and temperature), and therefore proactively schedule maintenance to avoid costly downtimes.
Such use cases appear to be the gateway to IoT adoption for many manufacturing companies. Proactive maintenance and real-time performance tweaks produce immediate cost savings, and help optimize output and efficiency. Manufacturers implementing IoT as part of the larger goal of EAM can finally start answering questions such as: what equipment will most likely need maintenance in the near future? How can we adjust workloads to optimize output while minimizing strain? What are the external factors most responsible for failures and how can we best control them?
2. Monetizing guaranteed performance: Taking asset performance management a step further, IoT won’t just prevent failures, but will also guarantee outcomes. Companies who primarily used IoT to monetize selling assets will be able to create new business models—forging contracts based on guaranteeing that their industrial assets will perform to a certain level.
Already, companies like GE are taking this approach with manufacturing customers. If executed well, this is a win-win for both buyer and seller. Equipment companies can use IoT to grow new revenue streams, while manufacturers who implement “smart” machinery can rest assured that their investments will produce tangible results. Profitability will be the key metric at the end of the day. IoT-ensured performance will directly correlate to cost savings and efficiency.
3. Customizable IoT solutions made possible by apps: When smartphones like the iPhone debuted, the hardware was impressive—touchscreens with integrated cameras and Internet capabilities held a lot of potential. But it wasn’t until app markets debuted that the real revolutionary effect of the smartphone was unlocked.
It’ll be similar with IoT. Hardware-wise, smart machinery with sensors and connectivity are impressive. But it will be the software that makes use of these features, and the vast data networks that underpin them, which truly unlock IoT potential in manufacturing.
Customized solutions will be the biggest boon. A manufacturer of smart phones and a manufacturer of laptops might share some industrial IoT equipment, but each manufacturing supply chain can be powered by a series of custom apps designed to optimize their specific processes.
The important element of this is that, just as smart phone apps exist on an established platform (iOS or android), custom IoT apps will exist on top of established data networks, which link disparate parts of the manufacturing supply chain. The advantage of this is that companies can create custom solutions without running into the problem of data silos and incompatibility between divisions and supply chain partners.
4. Advances in manufacturing automation: A lot of the initial focus on IoT is getting information out of equipment. The increased visibility provided by IoT means greater insight into processes, and better opportunities for efficiency and cost savings. But the most mature stage of IoT adoption will be a two-way flow of information.
Instead of just getting insights out of equipment, manufacturers will be able to push information back to them, changing settings, orders, operations, all securely and remotely. The linkage between information and control will allow the big data analytics and machine learning algorithms responsible for guaranteeing performance to adjust operations automatically based on real-time conditions.
IoT equipment will thus become the eyes, ears, and limbs of an intelligent manufacturing system that exists in the cloud of networks, big data, and machine learning. In the end, what we’ll have is complete feedback between real-time data, analytics, and control.
The progression toward these stages won’t necessarily be straightforward and linear. Different manufacturers will experiment with different aspects of this overall journey toward IoT maturity. But that end vision, of a fully automated loop running on real-world data, is the ultimate endgame of IoT in industry.
Here’s a scenario that showcases this feedback loop.
Based on certain data, a manufacturer notices that a piece of equipment is undergoing strain because it is producing one specific part at high volume. To avoid potential breakdown, the manufacturer decides to switch that part’s production to another piece of equipment, to ease the load.
A production change request comes into the manufacturing center digitally, via a supply chain network, after being approved by the necessary stakeholders. The change order then goes straight to the machine that will be taking on the new task. This machine coordinates with the original machine, located perhaps halfway around the world, to ensure a smooth handoff in production, ensuring that the total quantity of parts produced is as originally intended.
Big data analytics take into consideration the workload, stress, and capacity of both pieces of equipment, the final destination of the goods, and the adjusted transportation costs, delivery times, and overall profitability. Then, the entire system automatically optimizes operations, and begins altering production by pushing control changes right to the smart manufacturing equipment. Because all these operations are connected via a common data network, all the stakeholders involved are aware and up-to-date about the changes.
This is the sort of automated, revolutionary potential of IoT in industry. One of the crucial prerequisites to this end-to-end IoT-optimized production process is a data infrastructure that can connect all parts of a manufacturing supply chain together. That’s where the critical power of networks comes in. It’s no longer large technology companies like Google and Amazon that need to command vast data networks.
Everyone from manufacturers to retailers will need to have some level of connectivity linking their departments, vendors, and trading partners. Once that happens, piloting and experimenting with different IoT projects in smaller chunks becomes easier.
By 2025, the total global worth of IoT technology could reach up to $6.2 trillion. And the businesses who want a piece of that action are already putting their money where their hopes are.