Several challenges continue to limit the efficient implementation of Industrial IoT. Here's a look at some of the key IoT challenges and trends in manufacturing.
About two years ago, when McKinsey asked business leaders about being ready for industrial automation, 92 percent said their existing business models wouldn't survive digitalization. The conclusion then was that most companies weren't even prepared for the fourth industrial revolution or even in planning for the implementation of IoT solutions. The challenges that IoT faced were too many then, and the market was not just ready
to deal with them.
Fast forward to the pandemic year, Industrial IoT implementation has exploded into the market as businesses scrambled to minimize the number of people on factory floors and other workplaces for safety and higher efficiency. Japan recorded a 13 percent increase in the export of robotic solutions in the second quarter of this year. Chinese production of robots increased by 14 percent in August. The story is similar in regions like Eurozone as well.
Current IoT trends in manufacturing
A combination of COVID-19 concerns, a need for increased productivity, and a stricter risk/regulatory environment drive the demand now, encouraging companies to look beyond the problems in IoT. Along with more productivity, companies are also trying to diversify their services, offer additional products, and adopting better business models for growth.
Perhaps most importantly, the continued economic shift towards lower-cost sensors, connectivity, and cloud hosting costs will drive IoT growth in previously cost-prohibitive areas
There's a lot of discussion around AI/ML, Deep Learning, Fog/Edge, 5G, etc. enabling new and innovative use cases for IIoT. However, Rob Mesirow, Principal at PricewaterhouseCoopers, feels a more foundational trend will be a crucial driver in the explosion of IoT growth.
"Cost has been a huge barrier on the feasibility and scalability of IIoT deployments," Mesirow explained. "Device costs, cellular connectivity, and closed proprietary systems make it difficult to achieve ROI targets. We are seeing Low Power Wide Area Networks (LPWAN), such as the LoRa, Sigfox, NB-IoT, and LTE Cat-M as a key driver of lowering costs and ushering in IoT for the masses."
These technologies bring connectivity costs down to pennies a month, with lower module costs, better battery life, and increased coverage to enable use cases that never would have been achievable with "cellular economics."
"Imagine how many more things could be connected when connectivity costs drop from $10-$20 per month to $1 per month," Mesirow continued. "Driving down costs then justifies the business case and ROI. Customers are looking for solutions to solve business problems. These can be operational efficiencies, compliance with regulations, or safety/security."
The increased use of IIoT devices would generate more and more data. This data is at the crux of the fourth industrial revolution because insights gathered from this data help businesses make better decisions. While this is a critical advantage, Alex West, Senior Principal Analyst for Industrial Technology at OMDIA, points out that effectively collecting data is essential.
"There's no point investing in AI for analytics of data if you don't have the capabilities of collecting the data, to begin with," West said. "I think before companies look at the technology, they really need to do a bit of a self-assessment around where they are today. And what level are they at?"
There is a lot of interest now in AI, edge computing, and 5G. According to Intel, 5G is especially poised to transform industrial manufacturing, private, on-premise, and public applications. This represents one of the most significant opportunities to deploy next-generation wireless communications.
Again, this interest revolves around data, and it is where edge technology comes out as a winner. Intel points out that analysts expect that by 2023 more than 50 Terabytes of this data generated in various industries, including factories, would be processed at the edge. Processing at the edge removes many of the challenges that customers face concerning latency, bandwidth, reliability, security, and privacy.
Technological challenges of IoT in manufacturing
Despite better technology and the drive to make use of them, certain issues in IoT continue to persist. The need for rapid deployment that COVID-19 prompted has added to the problems. West listed three main challenges that customers face – legacy systems, cybersecurity, and skill development.
Retrofitting legacy equipment
An average manufacturing facility would have large machines that they have been using for about 15 years. These are machines that were never designed to be connected. Companies need to look at how and whether they should retrofit and upgrade these or work on other equipment to collect and analyze data as part of the whole connectivity trend.
Cybersecurity problems in IoT
Connectivity also brings in concerns and questions around cybersecurity, which is undoubtedly a big issue now
. Anywhere between a third and a half of manufacturing companies have experienced a cybersecurity breach in the last three years. And that's got a whole number of ramifications, loss of sensitive data, impact to production, damage to assets, etc. Dealing with cybersecurity problems in IoT is of paramount importance to several companies now.
Preparing the workforce
Although technology drives IIoT, it's also about the people. It's also about how you change your processes, how you bring people on that journey, and how you encourage your existing workforce to use some of these new technologies. A lot of the manufacturing workforce is aging. How do you promote this change of approach where some people have become very set in their ways? They trust what they do and could be skeptical of using new technologies.
Non-technical IoT challenges hurting the industry
Mesirow pointed out that end-user technological challenges such as battery life, cybersecurity, signal coverage, OTA updates, etc., have been well documented and regularly discussed. However, three significant themes usually emerge around IIoT deployments in addition to these technical challenges.
Industrial automation does call for significant capital investment, and for many companies, calculating the ROI on such a large amount is problematic. Mesirow suggests that before rolling out a solution, customers must answer, "what business problem am I trying to solve?". Next, do the benefits offset the solution's cost, and finally, where this sits on the list of priorities.
Lack of integration causing issues in IoT
Most of the critical values of IoT lie in the benefits they bring through integration. However, many IoT solutions are rolled out as "siloed point solutions," limiting their usefulness. Data is more powerful in the presence of other data, and there should be an overall IIoT Strategy/Capability roadmap that drives the selection of various use cases.
Need for operating model
The third and probably the most significant IoT challenge that customers must beat is coming up with an operating model that takes automation into account and sustaining it. These include processes and change management that incentivize employees to act on data and support procedures such as firmware updates and battery changes. Without a proper operating model, many implementations fail to achieve their business goals or "die on the vine" from lack of support.
The implementation of IoT in the manufacturing sector is growing as demand increases and solutions become more affordable. But customers still face several technological and non-technological challenges that the industry must come together to resolve. A key point to note is that technology's availability and affordability wouldn't matter if customers are not aware of best practices to get the most out of IoT implementation. A combination of education, integration with legacy systems, innovation in edge and deep learning, and open accessibility to developers would drive the growth of IIoT even higher.