How Private Networks and AI are Shaping the Future | Innovation at the Edge

In this enlightening episode of GXC’s Monday Meshup, host Allen Proithis, along with guests Patrick Parodi and Harish Nalinakshan from PwC, delves into the transformative world of private cellular communications and its impact on industries traditionally reliant on high levels of privacy and operational efficiency, such as oil and gas, mining, and utilities. The discussion navigates through the nuances of digital transformation, the pivotal role of connectivity in today’s fast-paced technological landscape, and the future of hybrid networks. As they explore the challenges and opportunities presented by the integration of artificial intelligence and machine learning in operational processes, the conversation sheds light on the critical need for security, the importance of unlocking data silos, and the promising horizon of private and hybrid cellular communications in driving business innovation and efficiency.

Industries that by design require private cellular communications are oil and gas, mining, utility industries, warehouses, logistics, and very large ports, such as seaports, airports, container ports, etc., where private wireless communications is a game changer simply because they get to automate a whole lot more.

Below is the transcript of the podcast.

[Allen Proithis] I’m your host, Allen Proithis. I’m thrilled to have two guests with such a deep understanding of the industry. I can’t wait to jump into our conversation. Patrick Parodi and Harish Nalinakshan are joining me today from PwC. Gentlemen, welcome to the show. 

[Patrick Parodi]  It’s great to be here.

[Harish Nalinakshan] Good morning. It’s great to be here as well. 

[Allen Proithis] You know, it’s funny. In this fast-paced world, just before we started, it was hard to believe a month had passed since MWC Barcelona. The industry is moving so quickly that everything’s just a blur.

But speaking of blurs, I know PwC is a very large place. Why don’t we ask you, gentlemen, to give us a brief background on your roles within the organization? Patrick, why don’t you start us off?

[Patrick Parodi]  Well, thanks, Alan, for this opportunity.

So, my name is Patrick Parodi. I’m a managing director here at PricewaterhouseCoopers (PwC). I’ve been with the organization for just over six years, spent 30 years in wireless, starting with wireless network engineering back in the early 90s, then made it up the stack, started doing mobile video, got acquired, was on the board of Shazam for six years, then did mobile advertising, and then found my way back to the United States about 12 years ago for a company called Light Squared, and then continued to work on connectivity, helped Facebook build their telecom infrastructure project to build their first wireless lab, then started doing some work with PricewaterhouseCoopers, particularly around LPWAN and other technologies that are now really becoming a core part of digital transformation and operations transformation for a lot of our clients. I’m happy to be here. 

[Allen Proithis] All right, Harish, why don’t we continue with the introductions? 

[Harish Nalinakshan] Thanks, Alan. By way of introduction, Harish Nalinakshan, I’m a partner within PwC’s tech media telecom space. I’ve been in the industry for about 25 years, 15 of which have been with PwC, 10 of which were actually in the industry working for several wireless and wireline companies, both in the U.S. and outside the U.S. So, seeing a lot of the space evolve over the last four generations of wireless, and probably one or two generations of wireline as well. What I do within PwC is help our clients across the board, primarily telco, but increasingly outside of telco, build, monetize, optimize, and sunset their networks.

Now, more than ever before, people are talking about custom networks, private networks, and private-public networks. The whole hybrid story is sort of the main message I came out of MWC with, and I’m excited to start talking more about how we bring a hybrid to the forefront here and get the most value out of the complexity of all the technologies we’re putting in there.

[Allen Proithis] Harish, you reminded me of the old phrase, one size fits most. And that’s not true very often. Everything has its place, but it can only do some jobs. But that’s a great perspective.

We’ve been talking about digital transformation for quite some time now. I know PwC plays a critical role in several different areas. Why don’t we start out with a little background on how you see PwC’s role in the larger digital transformation landscape?

[Harish Nalinakshan] I’ll take that. Patrick, you can add to what I add here. PwC, Patrick and I joke about private wireless communications. 

Now, increasingly, it’s private, public wireless communications. But in all seriousness, our role as a firm is to build trust in society and solve important problems. That’s the moniker that we go out to market with.

There’s deep meaning in those words. We build trust in society by instilling trust in how you get data, where you get data from, and how you look at the sequence of events of some things that have happened in the past. Solving important problems is really the other side of it, solving the issues of today, the core issues, communications challenges, sustainability, equality, et cetera, into tomorrow.

How we’re bringing that to market is really combining all these evolving technologies, particularly the digital ones, and using them to solve the issues of today and the future. 

[Patrick Parodi]  I echo what Harish is saying. PwC as a company has been around for 175 years, and we’ve been building trust with our clients and trying to find technology solutions that solve real problems for our clients and at the societal level. Examples of things we’ve deployed: this is the team I operate under, called Connected Solutions. We have an ecosystem of technology companies, such as GXE, where we orchestrate, deploy, and manage solutions for our clients that provide the right business outcomes.

We’ve done everything from rapid response buttons to housekeepers to nurses in large hospitals and large hospitality environments, but we also developed an enterprise-level contact tracing solution during COVID. My team leverages different types of technologies, primarily around extracting data from the physical world. In my team, we talk about digital transformation; it’s physical, digital transformation, or, as the CTO of a very large hospitality chain calls it, digital.

We leverage any connectivity solution or wireless technology. We’re big fans of private wireless, P5G, and LPWAN, and we provide that in our time series graph engine called SignalGraph. SignalGraph is an ML/AI-boosted analytics engine that analyzes data from the physical world, finds mechanisms to make things more efficient, effective, and safe, and provides the right business outcomes for our clients. So we’re seeing a huge boost in this notion of operational AI because we’re talking about new datasets.

Some of them already exist, but some of them are new. Most of the ones that already exist are siloed. And part of our mission is to unlock those silos and bring them under one unified, refederated data platform, applying the right machine learning and AI tools to solve problems.

So it’s pretty exciting stuff.  

[Allen Proithis] Well, Patrick, I love that word digital. I want to get that on the T-shirt.

It reflects the reality of the complexities of everyone talking about data being the new oil and the fuel for AI and ML, but you have to get it. And sometimes, that’s the hardest part. If you have it, how do you actually pull it all together to extract value out of it?

We see that, of course, is a big problem. We look at private cellular networks as one piece of the solution as well as the transport layer, but there’s so much more to it. What do you see as the biggest challenges?

When you get to that data level, you decide to solve a particular problem. What do you think about that data? What are the biggest challenges you’re seeing in putting all that together?

[Patrick Parodi]  Well, you know, it’s interesting. We just surveyed operations and supply chain officers, and we found that 69% of them feel that tech investments haven’t fully delivered the expected results. You could blame it on the technology or the investments, but in the end, it’s really about taking an approach that looks at things more holistically.

So a lot of organizations already have data, as I mentioned earlier, they’re quite often put in silos, whether it’s your building management system, whether it’s your IT infrastructure, things as simple as your, you know, for people coming ingress and egress from buildings and how you take the elevator inside of the building or your utility information, right? That’s all siloed. We’re seeing, particularly with the advent of these AI models, that some can leverage large language models. Still, some of them leverage straight algorithmic time series AI, where you’re using machine learning to understand an environment.

You start seeing correlations and seeing things that you wouldn’t be able to see otherwise if you didn’t unlock these operational silos. So that’s an area we’re driving many of our initiatives into. And it’s in places like large hospitals, for instance, or university campuses, or airports, or malls, or these large buildings in particular, where we focus a lot on in-building, particularly where we think that private wireless has an opportunity or at least large campuses, and unlocking that value, potentially from some of their existing infrastructure.

There’s a step change that’s required, and potentially, it is possible to leverage this data that already exists without the large infrastructure investments. Now, once you see the value of extracting this physical data, then the investment in this new type of private wireless infrastructure becomes much more apparent. 

[Patrick Parodi] That makes a lot of sense and is consistent with what we’re seeing.

Harish, you’re more customer-facing, perhaps, with some of the things you do on the front line. What do you see in terms of industries and use cases from your perspective? What do you think is moving faster than others?

One of the things we see all the time is that we know different technologies like this can be applied in many different ways in almost every vertical industry, but they tend to start faster in certain places. So, what are you seeing out there? 

[Patrick Parodi] It is a fairly wide set of things that you’re seeing out there, from the pace of change and so on.

For example, industries that by design require private cellular communications, such as oil and gas, mining, and utility industries, continue to double down in this space. They’re asking for more and more of this private wireless flavor. Of course, there’s this hybrid notion where they want to interface with the public networks where and when it makes sense.

The other group of industries that have now started to adopt this are the logistics industries, the warehouses, logistics, very large ports, for example, seaports, airports, container ports, et cetera, where private wireless communications is a game changer for them simply because they get to automate a whole lot more. And then some the masses of everybody else is starting to look at this and say, here’s an opportunity for us to actually move forward with the technology and where they can go. One of the inhibitors of where some of this is moving in the speed or the pace of change is around economics.

They’re still trying to determine the economics of their move. It makes sense in true remote settings like mining, oil and gas, and so on, but it’s starting to make more sense in your standard set of industries now.

There are a plethora of use cases, but oftentimes, what you start to see is that the first use case is usually underwater. But as you start stacking use cases one on top of the other, you soon find that you’re starting to get in the money somewhere around the second, third, or fourth use case, and the cases make a lot more economic sense.

Of course, the struggle always lies with that first use case. And then you’re also finding the richness of the device ecosystem. As you find again, oil and gas and whatnot, because of where they are and how they’ve been, a lot of it is they tend also to own some of their own spectrum.

There’s an ability to actually buy at scale for some of them or buy more expensive devices that are very bespoke. So those are the industries that are moving. And then we’re heartened by the fact that we’re actually seeing this wave come across.

If you had asked me this question a year or two ago, Alan, I would have said it was very slow. Last year, it started to turn. This year, we’re starting to get out of the gate more aggressively, and we’re making good strides in terms of more adoption across more industries.

It’s happening faster regionally as well. In parts of East Asia, for example, China and others, private networks are just going gangbusters. We’re seeing some of that pickup in Europe and the Middle East.

We’re starting to see that pickup here in America as well, but the pace of change is also fast. 

[Allen Proithis] It’s funny you mentioned logistics. I was recently at the big logistics show of the year, MODX, in Atlanta. And when you walk around that show, it feels like you’re in the factory in the Terminator movies where they make all the Terminators, the level of automation that’s going on at this point.

It’s funny because Wi-Fi certainly has a place. We always describe it as if there’s carpeting; Wi-Fi is great. But for the un-carpenters of the world, for more business-critical operations, you need something like private salaries and other technologies.

It’s funny because you have this incredible level of automation. To your point, a year ago, people were like, what’s private cellular? But today, I think they realize there’s a place for Wi-Fi, and it’s important, but it’s not sufficient.

That realization is that every nook and cranny of a factory or warehouse now, or campus for that matter, has to be connected securely. From what we’re seeing, that’s driving a lot more interest in a faster adoption on the private cellular side. The feedback we get, and I’ve met with five different CIOs of multibillion organizations in the last 60 days.

One of the recurring themes is security. So I asked you, gentlemen, what are you seeing regarding security concerns? We continue to see that it is becoming a bigger and bigger part of the decision-making process.

[Harish Nalinakshan] It is going to become a game-changer. So, as you start to deploy more technology, to Patrick’s point, as you start to collect more digital signals off of just about anything you want to measure, the requirements for security become even more important. Because now you’re truly digitizing mission-critical things on your factory floor, retail floor, assembly lines, et cetera.

Things that will make a difference around safety, compliance, etc., will all require an envelope of security far greater than you had in the past. And so the requirements become greater.

The security requirements around, if you do have a set of hybrid networks, so Wi-Fi, cellular, LPWAN, et cetera, you put them all together, you will still need to have a security blanket around the network of networks as well because you’re going to be collecting all this mission-critical data. So, it does become crucial. It will become even more critical as that data now winds its way from a sensor through a network, to a cloud, to an analytics engine, and ultimately in front of an executive’s desktop.

That entire data value chain, if you will, will need to be protected. And it’s going to have varying degrees of security depending on the criticality of the asset and what is being actually monitored, right? So things like biotelemetry, by the way, which is right around the corner.

I shouldn’t say around the corner; it’s already there today, but en masse, as it comes around the corner, you’re going to have even more varying layers of how much data, what data, how you capture it, where it gets streamed to, et cetera. 

[Patrick Parodi]  I would add to that notion of control as we look and see how this new operational data that’s part of the new world that we live in, particularly with AI and ML, will be controlled. The ability for your IT department to be up-skilled to understand these new types of wireless connectivity solutions and the new datasets that’s coming off of these assets, right?

Your people, your places, your things, and how these interact over time will require a new type of architecture as to where you host the data, how it gets computed, and then the decision sciences behind it, how these decisions are made. In some instances, they’re going to be made automatically without necessarily a human involved, and they go back into the physical world, particularly with the advances of robotics that will impact many of these different businesses. So we’ve been pushing a lot of folks. There are many reasons to continue to leverage the cloud, not least the effectiveness and the cost-effectiveness of hosting data. Still, I see a shift towards more of an edge compute model, potentially even on-prem models where the value of private wireless comes into play.

[Harish Nalinakshan] When you said control, Patrick, it triggered a thought, and you mentioned closed-loop control systems. Today, we’re talking about sensors, capturing data, processing it in the cloud, doing decision support, providing executive reporting, and so on.

But in the future, none of that’s actually going to show up at the desktop. It’s all going to happen at the edge, in the cloud, and you’re going to have closed-loop control systems effectively managing the issue or an incident right at the edge, closing it off, and reporting that as well.

That’s the sort of requirements that you’re going to see. And for that, Alan, it goes right back to the security layer around that, which is going to be quite intense. If you are talking about changing the direction of an assembly line, stopping or starting an assembly line, stopping or starting an AGV on a factory floor or in a warehouse, or closing the loop, those are going to require fairly robust security requirements just because of what you’re going to be able to do and impact.

[Allen Proithis] That makes a lot of sense. I was in a couple of hundred thousand foot cold storage facility for beef that, if ever completely full, would store over a billion dollars worth of meat at any given time. And you can imagine how operationally intensive that is with things coming in and out.

It was challenging not to get run over by the forklifts just taking a tour. But to your point, it has to have some level of more automated decision-making and problem-solving, or you start to immediately suffer operational deficiencies because it’s meant to be running at full blast all the time. The impacts, whether connectivity, data issues, or security issues, are almost instantaneously felt.

I think it’s really important to get that operational uptime using all these tools. But you know, we’re talking a lot about the data that you can collect with these networks. Then you have to do something with it, and you can do some really exciting things with it. But you know, this AI stuff has really jumped to a whole different level over the last year or two.

The companies sometimes struggle to say, okay, where do I get started, and how do I get started? Patrick, when you’re thinking about some of these solutions you’re working on, how do you help enterprises take that first big step to the next level by utilizing AI to use this data better? 

[Patrick Parodi] It’s a good question. You try to start small. You start with very specific use cases, and most investments are made based on solving a very specific problem. Whether it’s temperature monitoring for refrigeration inside a quick-service restaurant, for instance, we’ve got a very large deployment where different franchisees will have different types of equipment and different ages.

If you’ve got the ability to not only monitor the temperature but also how often the doors of the refrigerator open and close, or potentially even monitor the vibration of a particular motor that’s associated with refrigeration, you then have visibility into how these different devices are operating in their environments. Then, the correlation of those environments, meaning how often people come in and open the main doors of the restaurant, starts seeing patterns, and you can detect things that might not be normal and anomalies. And that’s where AI comes in.

The ability that you have, for instance, this type of machine is going to break over this period, can reduce the risks of, obviously, food board illnesses, but also the challenges of having to maintain and operate a business, changing equipment when it’s too early or changing it when it’s too late is bad for business. Being able to optimize your operations, leveraging this dataset, and leveraging AI and ML is the step change here, and showing it through actual live trials or proof of concepts is the best way to go for many of our clients. I think a lot of them fully understand the implications of these technologies.

And so we’re pretty excited to see how things are moving forward. 

[Patrick Parodi] That makes a lot of sense. Gentlemen, just to wrap us up here, I know that the pace of this stuff seems to be accelerating constantly.

What do you gentlemen see as the biggest trends over the next year through 2024? There’s a lot going on out there. And what do you think will be the biggest surprise or advancement over the next year?

Harish, do you want to jump in and start that? 

[Harish Nalinakshan] You’re going to continue to see Gen AI evolve. There’s been; I’m surprised we got this far into the conversation without mentioning Gen AI. Still, no, you will see more of that evolve as the language models change, compute changes, and more capabilities become available to us.

We’re going to start to see the ability to query your datasets, your large sets of datasets. You’re going to be able to enforce the closed-loop control systems that I mentioned and say, hey, stop that AGV or park that AGV, move this container to the left or move it to the right, 500 paces, and so on. Those sorts of things will be far more commonplace in the future.

We will see much greater adoption in the next 12 to 24 months of hybrid, private, and public cellular communications. And I say that and I also mean a network of networks where you’ll see a mix and mash of technologies come through because the use cases may start with one and evolve to another as the needs evolve for the business. You’ll see a lot of emphasis on data and the entire pipeline, the value chain of data from the source, and how it will be translated up to the finish.

Patrick mentioned converting physical to digital. I’m going to bolt on the conversion of digital to financial and mapping of digital to financial because, at the end of the day, it’s all got to make good business sense. 

[Allen Proithis] It makes a lot of sense there, especially the hybrid point, because every tool has a task that’s best optimized for, and it’s typically a combination of tools to get the whole job done.

But Patrick, what does your crystal ball tell you or your magic eight ball, whatever you prefer? 

[Patrick Parodi] I don’t necessarily have a crystal ball, but I have 30 years of experience in the wireless industry. Everybody talks about how Alexander Graham Bell said one day, every city will have one of these.

Our industry is notorious for things to take time, talking about M2M, now IoT for, I don’t know how many years, but then it’s always bigger than everybody expected. We’re at that inflection point, particularly with the advent of these new private wireless systems, these hybrid systems that Harish was referring to. That is going to start impacting businesses and impacting our lives, whether it’s you’re a college student, whether you’re a retail guest in a mall, or you’re a patient in a hospital. It will all generate data, which will be leveraged using artificial intelligence.

It’s exciting. I would add one other word besides fidgetal, which I’ve been using. You guys are welcome to coin it.

I think you’re going to start realizing that things happen just automatically in the physical world, as they have done online for quite some time because of the advent of this fusion of these new technologies. So it’s pretty exciting stuff, and I’m excited to see what happens next.

[Allen Proithis] Your predictions are spot on, so I’m very excited about that, Patrick.

The other thing about private 5G is that we’ve all seen a pop-up on our phones for the last five years, and we are wondering what’s different besides my battery going faster. The innovation to all the points everyone mentioned about the importance of gathering the data so you can do something with it. I see more innovation in the enterprise than the consumer side of 5G over the next year because the needs are stronger to justify it.

I see that happening. I’m slightly biased, but we’re seeing some of that already. But anyway, Patrick Parodi and Harish Nalinakshan, thank you so much for joining us.

It was great to have you here, and it was great to get your insights. Best wishes for a very successful twenty-four. 

[Patrick Parodi] Thanks, Alan.

[Harish Nalinakshan] Thank you for having us on.

GXC MeshUp | Key Takeaways

The podcast provides a comprehensive look into the evolution and impact of private cellular communications across various industries, with insights from Patrick Parodi and Harish Nalinakshan of PwC. Here are the key takeaways:

Private Cellular Networks as a Game Changer: Industries inherently requiring private communications, such as oil and gas, mining, and utilities, benefit significantly from private wireless communications. This technology enables extensive automation and operational efficiencies in large ports, warehouses, and logistics centers.

Digital Transformation and Connectivity: PwC’s role in digital transformation involves leveraging emerging technologies to solve critical problems and build trust in society. Their approach combines digital innovations with traditional operations to enhance operational AI, improve safety and efficiency, and deliver better business outcomes.

The Importance of Unlocking Data Silos: Many organizations already possess valuable data locked in silos. Integrating and analyzing this data using AI and ML can reveal insights and correlations that are otherwise hidden, leading to more informed decision-making and operational improvements.

Operational AI and Physical-Digital Integration: PwC emphasizes the concept of “digital” transformation, bridging physical operations with digital technologies. Utilizing private wireless, LPWAN, and other connectivity solutions enables real-time data analysis and insights, driving operational AI forward.

Security and Control in the Age of Connectivity: As more devices and operations become digitized, security becomes paramount. The discussion highlights the need for robust security measures to protect the data pipeline from sensor to analytics, especially as operations rely more on automated decision-making systems.

The Future of Private and Hybrid Networks: The narrative predicts increased adoption of private and hybrid cellular communications, reflecting a shift towards more sophisticated network solutions that cater to specific industry needs. This trend is expected to accelerate the digital transformation across more traditional sectors, enabling them to leverage the benefits of advanced connectivity and data analytics.

Challenges and Opportunities with AI Integration: Enterprises are encouraged to start small with AI, focusing on solving specific problems. The conversation underscores the potential of AI to transform operations by analyzing physical data, predicting equipment failures, and optimizing processes.

Looking Ahead: The discussion concludes with predictions about the continued evolution of AI, the importance of hybrid network solutions, and the shift from physical to digital to financial conversions. These advancements are expected to drive significant changes in how businesses operate and deliver value.

This podcast episode from GXC’s Monday Mashup provides a deep dive into the transformative power of private cellular networks and digital technologies across industries, highlighting the ongoing shift towards more connected, efficient, and secure operations.