Expert Review: 3rd Gen Intel® Xeon® Scalable Processors

“Move faster. Store more. Process everything.” Intel has aligned its portfolio to enable these three pillars, and with the recent launch of its 3rd Gen Intel® Xeon® Scalable processors, it is taking these ideas to new levels. So why does this matter?

With advances in multi-cloud computing, widespread IoT adoption, IT/OT integration, and 5G, data is increasingly generated everywhere, by everything. Indeed, the OpenFabrics Alliance anticipates a 600% increase in annual data creation over the period 2015-2024. Organizations that can collect, analyze, and share this data securely and intelligently are poised to outperform their competitors.

For me, the 3rd Gen Intel Xeon Scalable processors exemplify how the explosion of data can lead to rapid innovation. With more than 200,000 units shipped to major customers already, the latest Intel Xeon processors are available in a wide variety of servers, boards, and more.

The processor marks a clear generational uplift, with 46% increased performance—but more than this, it affords solutions to some of the key challenges impacting organizations today. My reflections on the leading advances and differentiators traverse AI, 5G, Security, and Tech for Good. Let’s consider these in turn.

Built-in AI Accelerators

There are many “flavors” of Artificial Intelligence, and I believe Intel’s leadership in CPU-oriented AI is especially important as this provides the flexibility to handle all workloads. But maximum performance requires AI-specific features—and the 3rd Gen Intel Xeon Scalable processors are the company’s first datacenter-class processors with built-in AI accelerators for superior performance.

The result? These processors deliver 1.5X more performance than other CPUs across 20 popular machine learning and deep learning workloads. And thanks to the open oneAPI standard, the new processors help engineers and data scientists build and deploy smarter models more widely, from proof-of-concept to production. This reduces issues around data and model waste, and enables Model Ops approaches.

Further still, the Intel® Deep Learning Boost (Intel® DL Boost) facilitates AI acceleration to significantly improve inference efficiency. One example of this in action is 3D digital face reconstruction for gaming.

Optimized for 5G Networks

5G is catalyzing a rapid transition to virtualization, vRAN, and cloud-native technologies, necessitating more agile network infrastructure with performance tailored to workloads. The 3rd Gen Intel Xeon Scalable processors deliver 62% more performance on a range of network and communication workloads, enabling lower latency, more rapid deployment, and higher throughput.

And with security a key focus for 5G, CoSPs can also reduce the performance impact of full data encryption and increase the performance of encryption-intensive workloads. This affords the capacity to extend existing 5G use cases and enable brand-new possibilities from edge to access, core, and cloud.

A Foundation of Security

There is a key emphasis on cybersecurity, which has never mattered more, with attacks and malware becoming a full commercial industry in recent years, and only accelerated by the pandemic. A range of options and security controls are provided to protect data and application code, notably Intel® Software Guard Extension (Intel® SGX), which uses a hardware-based Trusted Execution Environment (TEE) to help isolate and protect specific application code and data in memory.

This is especially significant to enable confidential computing and privacy-preserving techniques so that organizations can benefit from data sharing—without the data ever leaving its secure home—notably critical in sectors such as healthcare.

As Gartner research recently found, data and analytics leaders that share data externally attain 3x more measurable economic benefitthan those who do not. Facilitating collaboration with privacy preserving computation such as Intel SGX directly supports this opportunity. In addition, Intel® Crypto Acceleration can reduce the performance impact of full data encryption and increase the performance of encryption-intensive workloads.

What I especially appreciate here is not only the security advances but also the enablement of developers, reducing restrictions and supporting time to production and creativity. And finally, Total Memory Encryption (TME) supports the encryption of physical memory, preventing the removal and scraping of DIMMs. Impressive advancements across the board!

Tech for Good

And finally, I wholeheartedly believe leadership in technology goes beyond the tech itself, to the people behind it and the purpose to which it is applied—alongside organizational time, talent, and other resources that can be shared. Back in April 2020, Intel launched its own Pandemic Response Technology Initiative to support COVID relief with a commitment of $50 million to enable response and readiness, online learning, and a critical-needs innovation fund.

Fast-forward to April 2021 and the 3 Generation Xeon Scalable launch. Before the technology was even introduced, Tech for Good and community support took center stage with a renewed commitment of $20 million for the Intel Rise Technology Initiative. This is supporting projects and collaborations to address some of the world’s biggest challenges, from education to climate change.

Bringing this piece to a close, this demonstrates how the latest technology innovation leadership can advance not only sustained business transformation but can go further still, and help foster scalable social impact change, creating shared value. I could not agree more with Intel CEO Pat Gelsinger, who summed up the event: “Technology is Magic.”

Safety First with AI Vision Kit

While humans may have a short attention span, computer vision will never get tired. This matters because physical safety in the warehouse and on the factory floor is a 24/7 imperative. An increased demand on the supply chain due to the pandemic has put pressure on these operations. And workers who move freight and stock have one of the highest injury rates for all occupations, according to the Bureau of Labor Statistics.

When managers and other personnel are responsible for monitoring industrial settings, employee safety can be put at risk of human error. While technology may provide the answer, creating effective solutions may be easier said than done.

“Traditional solutions, such as lasers, were very complicated to install,” says Christian Lang, senior manager embedded solutions at Avnet Integrated, a global technology systems provider. “They also required a lot of effort to track.”

In addition, the older technology is set to overcorrect: “It shuts down a complete system if an employee entered a restricted area,” says Lang. “That can be dangerous.”

AI and CV Automates Warehouse Safety

A better approach combines AI and computer vision to monitor an area and deliver real-time analytics and insights to businesses. For example, Avnet’s AI Vision platform defines areas, detects where people should and should not be, and sends alerts (Figure 1). Warning people if they get too close to robots not only improves safety; it allows the machines to continue working uninterrupted by humans, boosting productivity.

Large warehouse with 3 people highlighted to show CV in action.
Figure 1. Employees in warehouse settings are automatically tracked for safety with computer vision. (Source: Avnet)

“If a risk is identified, the system can slow down the robots and machinery only in that area, saving time and increasing efficiency,” says Lang. “Every second lost in production is also a loss of money.”

In addition to monitoring surroundings, the AI-Vision solution comes preprogrammed with other dedicated real-world applications. Its people-counting feature, for example, checks for compliance and distancing rules. Automated real time headcounts from the cameras may reveal problems, such as only two workers in an area where they need more.

The motion detection feature can ensure that a person or machine is moving in the correct direction. The system can also analyze human behavior, creating a report, for example, if someone spends too much time in a parking lot. AI Vision also recognizes intrusion and has settings that can be used for crowd density control.

“In industrial environments, every functionality can be easily adapted to your scenario,” says David Meyer, embedded innovation marketing specialist for Avnet Integrated.

The system also includes GDPR-compliant recognition. Protecting privacy is a must, and edge solutions send only metadata, not personal information, to the cloud. “Currently the software is only detecting people in a stream as humans,” says Lang. “To ensure privacy, we do it in a way that the face is not visible in the video stream, and the raw data remains at the edge.”

In industrial environments, every functionality can be easily adapted to your scenario. @Avnet

Simplifying AI Technology

Edge applications can be difficult and time-consuming to develop, but ready-to-deploy platforms offer systems integrators more business opportunities because they don’t need specialized knowledge to deploy them. They can also set up systems faster – an important consideration for safety programs.

The solution includes the Avnet AI-Edge System Infinity AI-Cube, which is loaded with the Avnet software, plus a wireless touch keyboard, accessories, and two cameras. Footage is captured and analyzed at the edge. The system runs on Microsoft 10 IOT, uses Intel® processors, and the Intel® OpenVINO Toolkit. Furthermore, it can be connected to Microsoft Azure cloud via Avnet IOT Connect.

“You just plug it in, start the system, set up your peripherals, and it’s immediately ready to use,” adds Meyer. “It’s easy entry for every level of user.”

Smart Factory in Action

Avnet recently worked with a warehousing organization that had experienced accidents when workers entered restricted areas without proper equipment. The company implemented the AI-Vision platform to automate real-time monitoring. If an employee violated safety protocol, the solution sent automated alerts that managers could act upon immediately. As a result, the company reduced its accident rate, improved the security of its operations, and set a higher standard of workplace safety.

Because of its small size and adaptability, the solution can be easily scaled by adding cameras—supporting up to 64 cameras at the same time, giving managers insights to improve efficiency on the factory floor. SIs and customers can also get technical support from Avnet to customize the software.

As automation takes on more tasks, manufacturers will need to revise procedures to make sure humans and machines continue to work together as safely and efficiently as possible. An AI-based solution that easily accommodates new use cases will help them achieve that goal.

New Dimensions for Digital Signage

Digital signs are everywhere, but they communicate in only one direction. Imagine the new opportunities when signage becomes smart enough to deliver personalized, targeted content, which is interactive and highly relevant to its viewers. To keep evolving, digital signage needs to harness the power of audience measurement—enabling brands and marketers to optimize their content and campaigns based on performance data.

Achieving such two-way communication requires deep integration between computer vision-based insights and viewer engagement. But combining these complex technologies can pose a significant challenge for systems integrators and their end-customers.

But with innovative, no-code solutions, creating tailored digital experiences that respond to a target audience can be accomplished more easily.

Intuiface, provider of a no-code digital content analytic platform, and Sightcorp, an AI-based audience intelligence specialist, partnered to make this a reality. The outcome of this collaboration is the Audience Analytics Bundle, a comprehensive solution that runs on Intel® processor-based hardware.

Sightcorp delivers audience insights while Intuiface provides data capture and visualization capabilities for analytics. Combined, it’s a straightforward platform enabling SIs to build an interactive UI without having to learn to code, as shown in Video 1.

https://www.youtube.com/watch?v=zu-U0gqh5vQ

Video 1. Deliver targeted content and determine its effectiveness with metrics like views, impressions, and dwell time a digital display platform. (Sources: Intuiface and Sightcorp)

“In many #DigitalSignage environments, there’s value in understanding who is interacting with the content, and for how long. From there, we can gauge its effectiveness.” @geoffbessin @Intuiface

Targeted Messaging and Insightful Analytics

The solution measures impressions, views, and attention time on a per-content play basis. Screen engagement time also can be used for future data-driven decision-making. Real-time information about the audience is used to tailor content according to predefined rules based on demographics like age and gender. Plus, the technology supports a diverse set of use cases, from digital communication in retail, hospitality, and airports to corporate signage and education.

“In many digital signage environments, there’s value in understanding who is interacting with the content, and for how long,” says Geoffrey Bessin, CMO at Intuiface. “From there, we can gauge its effectiveness.”

Digital-signage network operators traditionally locate their displays in high-traffic areas like shopping malls, transit venues, museums, and other locations. But until now, there was very limited measurement happening around the number of actual viewers and passersby. Adding CV and real-time analytics to their existing digital-signage assets offers new opportunities for maximizing and monetizing audience reach and new revenue streams via impression-based sales models.

On the end-user side, brands and retailers can formulate KPI-driven project goals. The solution anonymously tracks the number of people that have been reached and how long they have engaged with the content.

“Advertisers benefit because they get accurate and reliable feedback on how their ads performed,” says Joyce Caradonna, CEO at Sightcorp. “They can use this information to optimize content to maximize impression counts within their target audience group for each campaign.”

Streamlined Integration With Digital Display Software

To take advantage of these new business opportunities, systems integrators (SIs) face new challenges. They need to integrate AI with an analytics-enabled content management system and incorporate them within their digital signs, which can be time-consuming.

“With our solution, we’re making it accessible for anybody,” says Bessin. “You don’t need a degree, write code, or access a complicated content distribution platform.”

The Audience Analytics Bundle allows operators, retailers, and the SIs who serve them to provision devices with everything needed to run interactive digital content on physical displays. Sightcorp’s DeepSight software gathers all data on the edge and sends the anonymized raw data to Intuiface to control the onscreen content. No personal data is stored—respecting and protecting an individual’s privacy.

Unlike traditional digital-signage platforms, creators are not constrained to a zoned approach. “Usually you have playlists for zones and you need to map it out that way,” Bessin explains. “With our solution, you control every single pixel, so the possibilities are unlimited.”

Once SIs create their content, they can customize the actions and triggers, and set specific data points to be collected. A trigger can be a certain time of day, viewers’ actions, or their demographics. This information is fed to the analytics dashboard, where stakeholders can identify trends and garner insights.

“The criteria could be very specific or very broad,” says Bessin. “Our system makes it easy for businesses to implement their own ideas, but they can also build templates to create a more prepackaged approach.”

Before adopting an analytics-based approach, SIs should work with customers to define their goals. Intuiface provides consulting services to help them evaluate their needs before they make the leap. “Determine the KPIs and whether they can be measured directly,” says Bessin. “If not, look for secondary elements that imply the key ones.”

Smart Signage at Grocery Stores

One example of the bundle in action is a supermarket chain in the U.S. that has deployed the solution in its stores to remind people to wear a mask before entering the store. The kiosks are also being used to display advertising content from local brands selling their items in the store. They can measure how many people are looking at an ad and directly correlate the impression data with sales of the displayed products.

“The supermarkets are seeing really good results from this, along with a new impression-based revenue stream from the advertising,” says Caradonna.

With a simplified implementation solution, Intuiface and Sightcorp customers stand to benefit from a host of new opportunities—from ROI reports for advertisers to analytics that track audience engagement for optimization. Smart signage changes the game in out-of-home digital advertising, which makes it a real moneymaker for operators, retailers, and integrators.

Telehealth: The Future of Patient Care

Telehealth has undergone a renaissance. While the concept has long held appeal, it’s been stuck in eternal “pilot purgatory,” plagued by a patchwork of healthcare technologies and clinicians weary of reimbursement hassles. But when COVID hit, adoption surged, providing a way to mitigate a dramatic fall in private practice and hospital revenue.

As providers saw firsthand the many benefits, including more-efficient visits and better outcomes for their patients, they have embraced telehealth as a vital avenue for care. “Acceptance of telehealth had been at a crawl, but COVID made it a sprint. Now we’ve crossed the chasm and burned the bridge behind us,” says eCare21 Chief Revenue Officer Gary Holderby.

A New Standard for Patient Monitoring

One area showing particular promise is remote patient monitoring (RPM). Nearly three-quarters of providers report they would like to continue offering chronic disease management to patients via telehealth even after COVID-19 protocols ease.

That’s where the eCare21 Virtual Care platform comes in,offering a streamlined path that addresses the challenges telehealth has previously faced. Its comprehensive integration features a single, intuitive user interface that eliminates data silos that make coordinated care challenging. And importantly, it provides an audit trail that offers billing capabilities that support seamless reimbursement.

As Holderby explains, the virtual health solution evolved from its initial use for senior care. The company began aggressively building increased functionalities into the platform to accommodate diverse patient needs, such as remotely checking blood pressure, managing prescriptions, and verifying people were following their doctor’s advice.  Soon the company had created a virtual cloud-based ecosystem that supported what had previously been disparate technologies.

“The interoperability wasn’t there—there were too many contracts and too many different technologies. To the industry it was like they were being sold a car one part at a time and expected to tie all those together on the back end,” Holderby says. “We knew that theprevious fragmentation was not sustainable, and healthcare providers were ready to evolve to a simplified solution.”

At its core, the communication platform gives providers a single source to view and act on critical inputs in real time. It addresses current gaps between caregivers, primary physicians, and specialists such as cardiologists by ensuring everyone has access to the same information. And it provides a secure repository for medical records that follow patients, eliminating a major hurdle in coordinating and managing care.

“Taken together, these capabilities allow providers to get back to the joy of their work because they are practicing at the top of their license.” @gholderby @eCare21

AI Technology Provides Ongoing Triage

Through the simple, comprehensive dashboard, patients are monitored 24/7. Every time a reading comes in, the AI capability evaluates it and generates an alert if it’s out of the range that was set. That notifies medical assistants or nurses who can act on it as needed, contacting providers only with an escalation they can’t manage. This constant stream of information also eliminates arbitrary requirements that patients undergo checkups on a predetermined timetable since any potential issues already will have been flagged.

“Taken together, these capabilities allow providers to get back to the joy of their work because they are practicing at the top of their license,” Holderby says.

Healthcare Tech: Flexibility at Its Foundation

Cloud-based and device-agnostic, the solution is easy to deploy with patients wearing any type of smart device, located anywhere in the world. “We’re flexible to the device, the clinical workflow, and the billing engine,” says Pete Stevenson, eCare21’s president and COO. “And that ability to upgrade or pivot as needed was vital, as the industry was wary to commit to one type of technology in case something better was introduced.”

Partnering with solutions providers has allowed the company to scale quickly. It realized early on that to impact more lives more quickly it needed to supplement the business model with systems integrators. “These companies are already out there, pulling together multiple solutions—it’s in their DNA and their mindset,” Holderby says. Intel® and Dell technology and partnerships have been essential to the company’s plug-and-play model.

Creating a Path for Value-Based Care

Reimbursement remains one of the most pressing problems in healthcare. While it has long operated on a fee-for-service model—where each service represents additional revenue—the future is value-based care, which incents providers to keep patients healthy and out of the emergency room. Yet providers have previously lacked a tool to monitor or help patients once they were discharged, a hurdle that eCare21 removes with its transparency into patients’ behavior and symptoms.

“As the system moves to an incentive-based payment model, providers are thirsty for solutions that offer insight into patients’ behavior at home,” says Holderby. ”It’s in the lived environment where they become noncompliant, and that is where costs escalate.” Keeping an eye on readings to allow early detection means that providers can step in at the first sign of an issue and intervene before a patient winds up in the hospital.

Reopening Public Spaces with AI

Jaume Portell public spaces

[Player]

From retail stores to vacation resorts, people are making their way back to public spaces. But they need to feel protected in doing so. AI-powered visual displays can give your customers a confident experience while enhancing business ROI at the same time—without breaking the bank.

In this podcast we explore the possibilities with Beabloo, a developer of solutions that personalize the customer experience. In our conversation with Jaume Portell, co-founder and CEO of Beabloo, we explore:

  • How to create spaces that are both safe and inviting
  • Ways to drive revenue through customer demographics analytics
  • What’s trending in retail, transportation, hospitality, and education
Apple Podcasts  Spotify  Google Podcasts  

Transcript

Jaume Portell: We think that next big wave, is making sure that everything that we see that can improve that customer experience is seen, is detected and then transform that into action, action to the customers and action to the managers of the stores.

Kenton Williston: That was Jaume Portell, Co-Founder & CEO at Beabloo. And I’m Kenton Williston, the editor-in-chief of insight dot tech. Every episode on the IoT Chat, I talk to industry experts about the technology and business trends that matter for developers, systems integrators, and end users. Today I’m talking to Jaume about the reopening of public spaces such as transportation hubs, retail establishments, hospitality venues, and schools. Beabloo has done some great work in the last year using multi-sensor Edge AI and intelligent signage to create environments that are both safe and inviting. I want to know what’s next for these spaces, and how to deploy the next generation of technology while protecting their budgets. So Jaume, welcome to the show.

Jaume Portell: Thank you very much, Kenton.

Kenton Williston: So, I know that you are a co-founder of Beabloo, and first thing I wanted to ask you was what exactly led you to start Beabloo, and how the company has evolved since you began it in 2008?

Jaume Portell: Well, actually, we started thinking that we could bring some of the intelligence that was by then deployed in the e-commerce sites, where e-commerce was controlling the message and measuring the impact of every single step in the process to brick-and-mortar stores. And the challenge was very interesting, because we had to bring the analytics first, and then also control the digital-message delivery in those physical locations. So we do this by using computer vision and other sensing technologies to understand how customers move, what they want, what they’re touching. And then we adapt the communication, the value proposition of the store, using signage, using electronic shelf labels, and also in sending hints to staff in the store to serve customers better.

Kenton Williston: Yeah, those are all really, really hot areas right now, so it’s great to see. You’ve been around since 2008, definitely a leader in that space. I am curious, though—I know there is a company called Metriplica that’s part of the Beabloo group. What is that, and what do they do?

Jaume Portell: Metriplica is the consulting arm of Beabloo. We realize that when you’re doing digital transformation of retailers, or airports, or banks, or retail banking, you are touching beasts that sell billions of dollars, right? So if you want to do a proper job, you need to prove the value you’re creating with your digital-transformation process. So it’s not about applying technology and that’s it; it’s applying technology with sense. And that we do measuring the before and after, creating control groups. And we do this with consulting firms, and we are collaborating nowadays with companies like Deloitte or Accenture. And Metriplica is our consulting firm, and we acquired them to bring artificial intelligence know-how to the company—to bring very brilliant data science to the company.

And we use their know-how to improve our algorithms, and at the same time, to deliver these direct success cases to the local market in Spain. And those success cases have help us to be known, to prove the outcome we generate with those digital-transformation processes, that they by themselves generate a positive ROI on the first year, and then help other systems integrators, or other large systems integrators, to replicate those success cases in other markets.

Kenton Williston: Got it. That totally makes sense, and leads me to the topic of today’s conversation, which is, of course, that the world was totally turned upside down last year with the pandemic. And there’s been a complete rethink about how public spaces—whether those are retail establishments, or airports, or whatever the case might be—need to run their operations. And so all these technologies you’ve been talking about—in terms of being able to observe the behavior of people in public spaces and create intelligent analysis of that—has become valuable for completely new reasons. And I’m wondering, from your perspective, what you see as some of the biggest challenges in public spaces as we move more towards reopening.

Jaume Portell: Well, first of all, we’ve been creating technology to improve customer experience. If you want to improve the life of someone or the experience of someone, you need to take care of that someone. That is what our technology does, is look at what people need and react to it. In times of pandemic, you want to deliver messages to customers to help them understand that they are working in a space that’s protected and properly clean and where measures of protections are properly taken. And that is what digital signage does in this context. We have analytics in that digital signage that senses how those messages are being understood by customers—if they are paying attention to it, if they are actually reacting positively to it. And we’ve seen that messages related to COVID protection, related to the new rules of the game—wearing masks or social distancing—this type of stuff is 300% more interesting to the audience than the value proposition of the retail space. So, extremely relevant: communicate the new rules of the game.

The next one is, are you wearing the mask if you’re walking inside? So we have trained our systems to detect masks on the faces and create real-time reaction to that, and our digital signage systems warn anyone trying to walk in a physical space where it’s regulated by mask protection, and in the case they are not protected, it warns them to not walk in. At the same time, physical distance is critical. We’ve trained the systems to sense physical distance between human beings, and we trigger alarms in the system if they are too close to each other. And that is also a real-time warning system that helps everyone to be aware that that’s a little dangerous. And on top of that, we can sense temperature using specialized cameras that are—on one feet of the camera detecting if you’re wearing a mask, on the other one detecting if you have fever potentially, and it triggers alarms to the staff in the store or in the bank branch or in the hospital or in the airport—to the staff to take care of that situation.

So technology that senses risk and communicates that to whoever can help to protect the staff, to protect the other users of that physical space. And we do this with computer vision. We also add some additional layers of artificial intelligence to clear noise from that data, to create heat maps that represent for the owners of that space where the interactions of risk are happening in that physical store, so that we can sense how to recreate the flows in that store to improve the security of people visiting. So there’s many things you can do. We’ve seen that the hardware in most of the cases that’s already there, it lacked some additional intelligence for the situation where we’ve been. We created that intelligence, and that same hardware now helps you sell more, but at the same time also protects your staff and your customers.

Kenton Williston: Yeah, so you’ve touched on a ton of interesting points here. And some of the things that really stand out to me—you’re talking about the idea of creating a space that, is it comfortable and inviting for customers? And of course, again, even what that means has changed a lot. But I think the implications around computer vision and AI—there’s a lot of sensitivities around that. So, here in the States, for example, there’s been a lot of, I might say, backlash—a lot of concern around how these technologies might be used, especially when it comes to sensitive things like racial profiling. And then, even globally there’s plenty of places where people have different sorts of opinions about mask mandates and things of this nature. So it seems to me like a double-edged sword. But I do know that you’ve done some really great case studies that really showcased the power of these technologies, and proved out—people really are having a positive reaction to this technology. So can you walk me through what you did there, and the results you saw?

Jaume Portell: So, basically, , we deployed our intelligent digital signage network with audience analytics. So we sense who looks at the content. When we say “who,” we mean the computer-vision systems see faces and take note of: this looks like a face of a man of this age. And it’s been looking at the content for 15 seconds. That is what we take note of. There’s no ID; there’s no way we can re-create an ID from that if the system sees the same face again 20 seconds later. It thinks it’s again, another face, different face, probably same age, same gender, but it has no clue that that is the same person. So no data-privacy issues at all. The system doesn’t record images; the system has been trained to count faces. It’s like you’re working in a physical space and someone is taking notes of 1, 2, 3, 4. So that’s what our system does. It’s very specialized to measure certain demographic characteristics, but they are anonymous, 100%.

We are GDPR compliant. It’s one of the toughest standards in the world for data privacy in Europe. And we apply it everywhere, with the goal of protecting the privacy of anyone that has been seen from any of those cameras, which doesn’t mean that the system cannot understand things that can be of certain risk. And this can detect two bodies too close to each other, or it can detect someone not wearing a mask, and it can trigger automatically a message saying, “Please remember that you have to wear a mask to be in this physical space.” But it’s not pointing to anyone, and it’s not recording any ID at all. So it’s a marketing system that makes physical spaces smarter, but it doesn’t trigger any concern about data privacy in customers at all.

Kenton Williston: Got it. That totally makes sense. In other words, it’s not like, “Hey, you—Jaume, you’re not wearing your mask. Demerits for you.” It’s more of a matter of, there’s a number of messages you could be displaying on these screens, anything from—let’s take the airport example again—directional information or flight updates or whatever, to all kinds of things that we’re specifically responding to, what was happening as people pass by—whether it’s people being too crowded together, or not wearing masks, or even if they’re revenue-driving things you wanted to do on that same display, like, “Hey, I see you’re of a certain demographic that might be interested in a new pair of headphones,” or whatever it could be. These are all things you could do, and all of these would be non-intrusive and not personalized in the sense of to a specific individual, but very meaningful to the audience, because they’re very responsive to what’s actually happening in real time.

Jaume Portell: Totally correct. We’ve done cases in airports where we’ve used the artificial intelligence technology to adapt the content based on—where is the next flight departing to? And that improved sales in duty free stores in a relevant way. And, at the same time, detecting if you have some staff in certain areas of the retail stores in the airport, or not, explains very much the conversion on those areas of products. And that is, again, another very valuable message, because it tells us how that customer experience is built up. If you went to a store in an airport and you end up buying a suit, and the reason why you did that was because you had half an hour waiting time.

That is very relevant information for the brand, very relevant information for the retailer, and very relevant information to improve the customer experience of the next guy like you—that if we detect someone walking in, probably it makes sense to have someone in the suit section to try to convert and create a better customer experience. So that is how the artificial intelligence that we are deploying in airports is enhancing, improving that customer experience—is taking care of the magic of what’s happening, and it’s controlling the messaging, and adapting it to the situation. The situation is the genders that are walking in, the families that might be walking in, the type of customers that might be walking in—still completely anonymous, we don’t know who that is, but we know that in thousands of times before, that type of customer wanted certain types of services or products.

So offering it improves the conversion, because it improve the customer experience. So that’s the magic of the systems. Right now, if you walk in to these spaces and I tell you how safe is the space, and how much we’re taking care of—cleaning and everything else—you will feel confident in walking in. If we tell you that we are counting every single individual that’s walking in and out, and we tell you: It’s now 10% full. So you know you’re free to walk in, it’s super safe—you will walk in way happier about that customer experience and your experience with that space will be more positive, and you will be more likely to purchase something from that location.

Kenton Williston: So, I think one of the most interesting things about everything you just described here is, it really highlights how you’re not only implementing AI technologies on the spot at the Edge to say, Who’s coming in? What kind of person or group of people is this? What are they doing in terms of masking, social distancing, etc., but also that added layer of intelligence, like, okay, if there’s a flight coming soon, where is it going to? What kind of sales history have we had in the past for these kinds of demographics? And I think that gets to the point you were making about that higher-level AI. It’s not just the on-the-spot recognizing the demographics and the social conditions, but also taking that to the next level and saying, “Okay, we have history now that we’ve developed, and we can start bringing some even higher-level intelligence to what actions we should take for this particular scenario.” And that’s really powerful. I love that.

I do want to talk, though, about some of the other sectors. So, everything we’ve talked about so far has been in the transportation sector, and really great to hear how you’re enabling transportation hubs to reopen, because I think that’s really critical to getting our world back on solid footing. But then there is the—a world more broadly of retail, and similar customer-facing establishments like banks and food service establishments. And what do you see as some of the big challenges there, and how are you addressing those?

Jaume Portell: So, retail banking is in serious transformation all over the world. They’re trying to help us all use the banking systems online, and they keep a certain level of branches open for higher-level face-to-face services. Their goal is to improve the customer experience in those locations, and help us in the journey of financing our dreams while we’re in those locations, right? So they want to explain to us that we can finance the university of our children, that we can get a system or a loan to pay for our next trip to Maldives, or whatever. So they want to help us on that front. And that requires communication, and it requires customer experience. If our visit to the bank is a pain, then we don’t want to come back, and we might end up doing that by ourselves somewhere else.

So they want to be our partner to make our life better and easier in the area of purchasing those dreams of our life—our wedding or the university or something else. So if you want to do that, you have to have a great experience in the point of sale. And that requires communication, it requires music, it requires sensing waiting times. It requires understanding the quality of service you’re delivering by asking your customers with digital customer-satisfaction surveys, and also measuring with analytics their waiting times, and when and how they are enjoying the experience. And that is a lot about sensing, and it’s also a lot about understanding what happened when you implemented certain campaigns to flood the traffic in branches. So queuing systems or scheduling visits to your bank branch is having an impact on that.

Measuring the waiting times in banks can confirm that computer vision is key to understand waiting times. You need to connect one camera to the next one and the next one, and follow customers when they walk in, so that you know who is getting in, who’s getting out. Not who with an ID, but who as to count how many seconds each of them had in the branch. And that is very relevant to measure the customer satisfaction, and that is very relevant to build loyalty in the customer. So banking needs sensing, needs digital transformation of the stores with computer vision, with those screens that explain the value proposition of the bank, the values of the bank, and it needs to be well connected with a queue-management system so that everyone feels informed of when they will be served.

So all this is perfectly possible with technologies like the one Beabloo is doing, and it enhance that customer experience a great deal. With all those technologies understanding where are the zones of risk of COVID-19. Or which are messages that are getting higher impact in your customers when they are in the store is another usage, very relevant in these COVID-19 times. But, actually, we sense and we see how valuable it is for customers, and how valuable it is also for the staff of the banks, because they also enjoy that level of information. They like to know that the occupation of banking is within the limits, that the distance is most of the time followed properly, and the measures of protection are working well.

Kenton Williston: So, Jaume, I have to tell you, this is a very relatable set of examples. I’m right now just about to close on buying a house. And I’ve been working with different financial institutions in order to do this. And it’s funny—some of the folks I’m working with have been fully remote, and some people I have worked with in person. And it’s wonderful to be able to do things remotely, but I have to say, the experience of working with people face-to-face, or mask-to-mask as the case might be, has really been so much better. And I think there’s clearly a need for institutions like banks to be able to provide an inviting environment for people to come into—to provide an environment where they feel like they’re getting great service, where it’s very enjoyable. Because, I have to say, the people that I’ve worked with in person I feel much more connected to and attached to, and I can definitely say from my own experience that I’m much more likely to go back there and use their services again, versus the people who are remote and are just an email address.

Jaume Portell: Yeah, human contact happened to exist before even the language existed, right? So we are rooted to looking at our eyes and trusting each other. And that is extremely valuable in retail, and it is extremely valuable in banking. Sometimes you need someone to trust, and that’s the value of local services. You can look at their eyes, and you can understand that they are doing the right thing. And that is unique, and it’s not transferable to an e-commerce website. And that is the value of human beings serving human beings in a direct mode. And that is measurable with computer vision. We can measure the staff context with customers, and how much value is that creating for brands in physical locations? And that is revealing the value of those interactions, of that local service that cannot be changed by an e-commerce transaction.

Kenton Williston: Yeah, absolutely. Totally agree. So, another thing that I have to say I have a very personal connection to is the education sector. So, my daughter right now is just in the other room, in our very small apartment, which is why I’m currently looking at buying a house—we’re too close together. So, again, I’m very grateful that she’s been able to continue her education remotely. But I’m very excited for her to get back to school and for us not to be so close together all the time. So, what do you see in the education sector as being some of the ways this technology can help schools reopen?

Jaume Portell: Yeah, well, digital communication in schools has been there for a while—the intranets, the local content delivery—mix hybrid systems, where part of it is online, part of it is face-to-face. I think there’s a lot of value to be offered in the interaction between the community and the organization. So, the professors talking to the organization, the students talking to the university, and doing all transactions by interacting with kiosks—by interacting with physical, digital displays that can talk to them and can understand them way better than their traditional web interfaces. And I think there’s a big, big opportunity there in voice and in natural language processing systems that understand what the customer is saying, what their intent is, and navigate them through the complicated systems of the schools or the universities in order to get information on how you’ve done, and which is the next step you need to do in order to graduate, or get to the new course, or whatever.

And navigating through those information systems is a very, very complex challenge that, when articulated through natural language, is way easier. And that is where universities will save a lot of time, create a better customer experience—in that case, customers are professors, and the whole community and students as well—and serve them better with information they need to do the work, or to just complete their tasks.

Kenton Williston: Yeah, I think this is a great example—what you were saying earlier about different kinds of sensors that can be incorporated into these systems. It’s the vision, it’s the thermal sensors, it’s audio sensors—probably other things as well. And I think we put these different technologies together, along with the higher-level intelligence—whether it’s knowledge about students’ needs, or travelers’ needs, or people in banks—you can begin to come up with some very, very interesting use cases. One last area that I want to touch on, though, is, I think, probably the most challenging if I’m thinking from a health and safety perspective. And that is in the space of things like hospitality and event venues and entertainment complexes—where the whole point of these facilities is to get people together and have them stay together for some period of time. And of course, this is very counter to how the world has worked for the last little bit. So what do you see as being some of the most important trends in those sorts of applications?

Jaume Portell: In computing, occupation is extremely relevant. Understanding when these spaces are full of people or not, is critical to cleaning them when they are not, and getting them ready for the next set of users. So, you with your family are in a resort. You want to go to—you want to go to the swimming pool. You would love to have your hamacas and stay there to have a drink. There’s a table—you don’t know if that’s clean or not. So you want to know if that’s clean. Having computer-vision systems observing the space to make sure that after someone uses it, someone went from the staff to clean it up, will give you the confidence that that place is right. So that’s a simple use case, but it is very relevant. You want to know that the space is available. Interaction Care, our system of understanding if a physical space has too many people or not will tell you that.

And you want to know if it has been cleaned up for your usage and the safe usage of your family. And that can be told by a system that’s observing the occupation, and, later on, it’s observing if the cleaning staff went there to take care of it. So you can see the tables—all of them empty, and you can see that nine out of ten are actually clean. But there’s one that is not yet cleaned. So that is, in form, in a digital signage system you know where to go, because there is a computer-vision system taking care of that. So it’s taking care of the invisible enemy of COVID-19. Because we human beings bring it here or there, but we human beings can make it clear for the next usage. So that is how, in hotels or resorts, you could use this type of technology.

Kenton Williston: So I’m going to take just a second here—I want to do a time check because we’ve got about eight minutes left in the hour. I can go a little bit longer, and I think if you’re available for maybe another 10, 15 minutes past the hour, we could get all the way to the end of our questions. Let me just real quickly send my colleagues a note, because I’ll be a little bit late to the next meeting. Okay, perfect. So, yeah, that totally makes sense. And I’m curious—do you foresee that there might be government regulations in the future around this sort of cleaning? I’m sorry, just a second here—just getting a poke from my colleagues.

Okay. So I’ll start my question over again. So, yeah, that totally makes sense to me. I can see how that would bring a lot of comfort to guests. And I’m also curious—do you foresee the possibility of government regulations around some of these cleaning protocols that you could help organizations meet? Or do you think this will be more of an informal, based-on-customer-needs scenario?

Jaume Portell: I think there’s two areas, two folds. One is what governments are enforcing you to do—or to restaurants, to hotels. This is happening in Europe—there are regulations, you need to follow them. But the most important one is the signature of quality of that space. So, actually, hotels and restaurants want to do it even better and communicate it like that—proving that they are doing it like that, so that everyone feels safe. And I think this is the point—it’s not about doing what the government says, it’s about doing what your customers will appreciate you doing, and communicating it, and making sure everyone knows that you’re doing the right thing, because that makes everyone feel safe. “Oh, no, please don’t sit there—it hasn’t been cleaned yet. We will clean it, then you can use it safely.” And that is the point, is you want to feel that they are taking care of you. And the best regulations are the ones self-imposed by those bars and restaurants and hotels to take care of their customers in their own way.

Kenton Williston: Got it. And the other big question I have is the cost question. So, it’s obvious to me how beneficial these technologies are. But I think there’s always a question of—especially given how tight budgets have been for a lot of the industries we’ve been talking about—how affordable these technologies are.

Jaume Portell: Well, actually it is surprisingly affordable, because how many retail spaces do you know that have security cameras, CCTV cameras? Many, right? Or banking—more of them, or hotels, many of them too. So those are the devices that they need to take care of their customers. How many of them have digital screens? Many of them have. And the thing is, the only issue is not the hardware—it’s the usage of the value that hardware is generating to actually make the places smarter and take care of customers. And that means connecting those streams from the cameras to a computer—an Intel computer that can get the information out of those streams, make sense out of it, and react in real time using the digital signage that anyway was available in the front door of the location. So what we’re talking about here is injecting intelligence in the hardware that is already in those physical spaces.

So we are talking about software deployment. It’s easy, it doesn’t require much installation, and it’s creating value by itself immediately, and, actually, it usually pays off for the hardware as well. Even if it was already there, right? So these deployments can be started from scratch, they pay for themselves, and they bring a positive return of investment, even taking care of the hardware cost. But if you already have digital screens, CCTV cameras, networking devices—you can use all of them to create more value with Beabloo Active Customer Intelligence Suite, that will sense with the cameras, sense with the Wi-Fi access points, understand what’s going on, and trigger the right answer to your digital signage systems, and actually analyze also the sellout of your stores of your cafeteria of your retail space to understand what’s working, what’s not working in your digital campaigns, and improve the selection of them based on the hour of the day, based on the day of the week, to increase your customer service perception and their conversion to sales.

Kenton Williston: Yeah, so, in other words, what you’re saying is most of the infrastructure is already there, and it’s really just a matter of making sure you have somewhere in your system—some Intel-based hardware that can run the Beabloo software, and then you’re good to go. And, like you said, yes, there’s some cost to installing these things, but they very quickly pay for themselves both in the ephemeral, hard-to-measure, customer experience, and very, very directly in terms of helping you optimize the digital advertising that you were doing anyways, and making that much more effective.

Jaume Portell: That’s absolutely right. I couldn’t explain it better. I will record this one, then I’ll make sure I can use it like that. Very well phrased—thank you.

Kenton Williston: Perfect. Well, so I want to think a little bit about where we’re going next, and what organizations who manage all these different kinds of public spaces should be thinking about as we move forward into the next, post-pandemic era.

Jaume Portell: The first thing is, when you start thinking about improving customer experience or taking care of your customers in a situation like the one we are in right now, you are sensing customers and giving them what they want as soon as possible. This is exactly what you want to do when the pandemic is over. You want to keep serving them and taking care of your customers as much as possible. So part of it is real-time reaction to it. Part of it is making sure that the stores and the physical spaces give the information the audience needs at any given point of time. So, real-time preparation for the scene and for the context—every hour of the day, every store, every space differently. But then the next one is to prepare the staff in the store so that they are ready to serve customers better when the day starts. So all these systems that we’ve been discussing are sensing people passing by, people walking in, people buying stuff—they know what the customers are buying, when they are buying it, so they can feed in an artificial intelligence engine.

In this case, it doesn’t need to be run at the Edge. Beabloo runs that on Intel computers, but in the cloud—can train a machine learning system to understand what will happen tomorrow. Today, this morning, was a sunny day in Barcelona. A sunny day has a completely different behavior for retail environment than a rainy day. And considering that today is Thursday, the purchase pattern of customers in that particular store is different than tomorrow, which is Friday, and it’s the day before the weekend. So these things are seen, perceived, and understood by machine learning algorithms that can explain it, when the day starts, to the staff in the retail store. And they can tell them, “Look, today’s Friday, and this weekend there’s going to be good weather.” So most of our customers will buy barbecue stuff from the supermarket.

“So make sure that we are replenishing that area as often as possible, because otherwise we might not serve them as well as we would like to.” That simple. I mean—and this can be sent as a video message to the staff in the store in the early morning, so that they understand what will happen and how to serve their customers better. So the intelligence is there also to empower staff to provide a better customer experience. And that is the next big thing—is sharing that intelligence with everyone, sharing that intelligence with the customer, sharing that intelligence with the staff. So if we’re counting how many people are in a given store at a given point of time, we can warn someone walking by that it’s not safe to walk in, or it actually is super safe to walk in.

We’re sharing the intelligence with the customers. If we tell the store manager that, today, they will be selling a lot of ice cream, because there will be a peak of temperature at noon, and please make sure that you have the right level of stock because we want to serve our customers better. This is using the intelligence of the store to help the staff to do their work better. So it’s about sharing the intelligence with everyone that can increase the customer experience. And that’s the point. And we think that the next big wave is making sure that the intelligence is collected from the hardware as much as possible, so that everything that we see that can improve that customer experience is seen, is the tech that is analyzed, and then transform that into action—action to the customers and action to the managers of the stores.

Kenton Williston: Perfect. Well, I look forward to eating ice cream together on a sunny day soon.

Jaume Portell: Me too.

Kenton Williston: So with that, I’d just like to say, thank you so much for joining us today.

Jaume Portell: Thank you for this wonderful time. Your questions have been very interesting, very enlightening for me too. It’s always very nice to have these types of discussions.

Kenton Williston: And thanks to our listeners for joining us. To keep up with the latest from Beabloo, follow them on Twitter at beabloo, that’s b e a b l o o. And, if you enjoyed listening, please support us by subscribing and rating us on your favorite podcast app. This has been the IoT Chat. We’ll be back next time with more ideas from industry leaders at the forefront of IoT design.

AI and CV Transform Digital Interactions

As online commerce shifts into overdrive, shoppers need a compelling reason to trek to physical stores. Interactive displays could do the trick. Picture a giant screen where people can play games, watch a parade of psychedelic-colored blossoms drift by, or snap a photo of their kids petting a virtual leopard.

Scenarios like these give shoppers an experience they won’t soon forget, while the technology behind them lets retailers capture valuable customer information. Although they’re a great draw, interactive installations are often expensive for retailers to install and difficult to manage.

But new solutions are emerging to lower costs and make them easier to use.

Unifying Digital Display Technology

Interactive installations include a wide array of equipment. Retailers may set up several screens equipped with computer vision cameras, which detect shoppers’ age and gender and show appropriate real-time promotions. They may also use QR codes, or place RFID or NFC sensors at point-of-sale exhibits, triggering product information to be shown when a customer lifts an item from the shelf.

“If you pick up a sneaker, you can learn about its features and find out, for example, if it provides traction in the rain,” says Julià Carboneras Girgas, chief marketing officer of interactive software provider Broox. “If you pick up two sneakers, you can compare their features.”

But what’s fun for shoppers is a headache for retailers and systems integrators. The hardware and software needed to deploy interactive systems are produced by different manufacturers, requiring multiple leasing or purchase contracts. Setting it all up and managing it is time-consuming and expensive. In addition, data from the hardware is sent to separate platforms that don’t communicate with one another – or with the marketing systems that retailers use.

“Retailers have dozens of platforms and dashboards to manage. You cannot mix the RFID data with the computer vision data, and you cannot put any of it into Marketo or Salesforce or Google Analytics,” says Albert Minguell Colomé, founder of Broox. “All our systems integrators and retailer partners were telling us, “Please give us something we can afford and manage,” adds Carboneras.

In response, Broox developed a “phygital” software platform that unites the in-person shopping experience with the digital world (Figure 1).

Digital display with running shoes nearby, and analytics examples
Figure 1. Broox integrates information from interactive technology hardware components and displays it on a single platform. (Source: Broox Technologies)

Broox builds its software infrastructure with the Intel® OpenVINO Toolkit, which speeds integration of AI capabilities and enables cameras, sensors, and other interactive technology to speak the same language. Hardware components send data feeds to a single platform that’s easy for retailers to manage. Sensitive customer information such as facial characteristics is processed on-site and with only metadata sent to the platform’s cloud.

The Broox platform also integrates with marketing analytics software, allowing retailers to do a deep dive into customer preferences and grow profits with cross-channel campaigns. Retailers who personalize the customer experience across physical and digital channels can achieve a 5% to 15% revenue increase, according to McKinsey.

“By using this platform, retailers can focus on creating content and forget about the technology that scares them,” Minguell says.

“By using this platform, retailers can focus on creating content and forget about the technology that scares them.” Albert Minguell Colomé, Broox Founder

Lowering the Cost of AI Technologies

The Broox SaaS payment model helps allay another retailer fear—excessive costs.

“We work with customers to determine their needs,” Carboneras says. “We’ll say, ‘For this idea, you’ll need the camera and the media player, and here is the monthly cost of that solution.’”

The platform also lowers expenses by working with Intel® NUC PCs. “Affordable PCs with a small footprint are exactly what our systems integrators are demanding. If software only runs on laptops that cost a thousand euros, nobody can roll it out in a hundred stores with 200 displays because it’s super-expensive,” Minguell explains.

Retail Analytics Offer Choices

Adding online options to point-of-sale transactions provides flexibility for customers, who are increasingly using stores as showrooms.

“Teenagers are famous for saying, ‘Let’s go to the shopping mall to buy something online,’” Carboneras says.

When a customer lifts a product from a smart shelf, scans information with their phones, but then doesn’t buy it, the retailer can send targeted promotions and track conversions through cookies. And someone who wants an item that’s not available in the store can click to have it delivered to their home the next day.

“Retailers can see that the online purchase was initiated in a store and learn the performance of that point of sale,” Minguell says. Broox is currently working on technology that will allow customers to purchase both in-store and ecommerce items in a single transaction.

Digital Experiences Revive a Mall

A shopping mall in Andorra illustrates how interactive experiences can spark visitor interest and boost profits. The mall connected existing security cameras and indoor and outdoor screens to the Broox platform. A new computer vision system captured visitor counts and customer demographics as screens displayed augmented-reality images of animals. The indoor screen also flashed a QR code (Video 1).

Video 1. A shopping mall entices visitors and captures information with unifying digital display technology. (Source: Broox Technologies)

People who downloaded the code to their mobile phones could snap photos as their kids frolicked with the animals. During the pandemic Christmas season, the mall switched the display to show Santa, so kids could have virtual visits.

“Previously, the screens were displaying boring, static images,” Minguell says. “The platform allowed them to create magical content and gain some nice analytics about results.”

Retailers gained a trove of customer information. In three weeks, the technology attracted 30,000 interactions. It increased mall entry by 7% and captured 4,500 new email addresses, becoming the best-performing campaign in the mall’s history. “And that was during the pandemic,” Minguell says.

As online shopping grows, Minguell expects to see fewer traditional and more experiential flagship stores. “Stores will become places where people don’t necessarily go to buy products, but to have a playful, enjoyable experience,” he says. “You can learn a lot from your customers when they interact with your brand.”

Must-See Tech at Hannover Messe Digital Edition

Hannover Messe is one of the world’s largest events, attracting a diverse international audience, and this year it’s going all digital. With more than 1,500 program items—covering the hottest trends in the industry, energy, and logistics segments—it’s an ideal platform for networking and innovation in this era of industrial transformation.

Here’s our roundup of must-see demos and sessions at this year’s virtual event, HANNOVER MESSE Digital Edition, taking place April 12–16.

Explore New Opportunities

As digital transformation expands, innovators are exploring opportunities to apply new technologies in manufacturing and other industrial environments. Autonomous mobile robots take a big step toward Industry 5.0. You can see them in action at a virtual demo showcasing the first AMR powered by the latest Intel® Edge Controls for Industrial, in partnership with SYMG, ASUS, and IEI.

See how SMBs can use open-source IIoT solutions to build new business opportunities—don’t miss Unleash the full potential of Industry 4.0, a livestream event hosted by Intel® in partnership with MicroStep. The session highlights IndustryFusion, an open-source IIoT connectivity solution for smart products and smart factories that enables SMBs to get involved in new business models like production sharing, pay-per-use machine usage, and CO₂-neutrality monitoring.

Check out the AI-driven factory of the future, a ThingTrax product demo that shows how any factory can become a smart manufacturing hub in less than 30 days. Cameras and sensors capture data on machines, people, and processes, while its AI platform converts it into actionable insights.

Also, take a look at AI and Real-Time Controller for Next-Gen Autonomous Industrial Control, a testbed demo environment that uses AI and real-time controls in the factory to help ensure safe working conditions. This tech is powered by Intel® Edge Insights for Industrial and Intel® Edge Controls for Industrial and developed in partnership with Hitachi Vantara.

Here’s our roundup of must-see demos and sessions at this year’s virtual event, @hannover_messe Digital Edition, taking place April 12–16.

Get Smart on the Factory Floor

Take a deep dive into why the edge is the key ingredient of any digital platform. See how TSN and OPC-UA run a smart factory on an edge controller with cloud connectivity—powered by Intel tech—enabling accelerated intelligent vision processing and control at the edge. Catch the livestream, presented with partner EXOR.

Smart Factories in Action

In the world of Industry 4.0, the emergence of open, tightly scheduled industrial communication is allowing smart factories to respond in real time, transforming manufacturing and boosting productivity. Join this Smart Factory demo to see how Intel® FPGAs help companies overcome key challenges in smart-factory implementation.

And one of the biggest challenges on the factory floor is quality control, an important step in manufacturing and factory pipelines. This defect inspection demo shows how the Corerain Intel FPGA-based AI accelerator can fast-track industrial video defect inspection. You’ll see how a smart hardware solution can perform a complex, automated visual inspection in the manufacturing process.

Register here for Hannover Messe Digital Edition to check out these events and demos, along with many more. We look forward to (virtually) seeing you there!

Amarillo Transforms Public Safety with HCI

Organizations need agile, reliable systems now more than ever for mission-critical activities. Traditional datacenters can have silos that impede workflow, which is why more systems integrators and IT leaders are turning to hyperconverged infrastructure (HCI) as the solution of choice. With a future-proof structure, HCI consolidates compute, storage, and networking functions into a single system to simplify data management, especially in virtualized settings.

The shift can be substantial, especially when old systems need updating. The City of Amarillo, Texas learned this firsthand when it hired Richard Gagnon to be its CIO in 2016. Amarillo’s public safety operations had been run from a large, flat network, Layer 3 firewalls, seven types of storage, and outdated laptops. The infrastructure was complicated, costly to maintain, and vulnerable to security hacks. Worse yet, it was slow, which negatively impacted vital service response times, including firefighters, police officers, and emergency medical technicians. An overhaul was overdue.

Gagnon, a seasoned global tech executive, chose Dell Technologies as its IT partner and transitioned to Dell EMC VxRail HCI. The solution integrated with the city’s existing VMware Horizon virtual desktop infrastructure (VDI) mobile network.

“Being a private-sector guy, I faced a lot of internal skepticism about the changes needed and my urgency to enact them,” Gagnon says. “But I have to hand it to my IT team. Keeping all this legacy infrastructure operating meant they had to really stretch their resourcefulness and creativity.”

Consolidation Maximizes Performance and Lowers Costs

Powered by Dell EMC PowerEdge servers, VMware HCI software, and VxRail HCI System Software (Figure 1), the quick-to-deploy solution consolidated Amarillo’s seven storage platforms onto a single appliance. Twenty-nine nodes in various cluster configurations took over a variety of workloads, scaling to maintain performance levels with software-driven automation and lifecycle management. And VxRail leveraged Intel® XEON® processors and Intel® Optane persistent memory to maximize performance at a lower cost.

VxRail HCI software and hardware elements
Figure 1. The VxRail system stack makes updating software as easy as a single click, helping eliminate downtime and protect resources. (Source: Dell Technologies)

Adopting a modern data management solution like Dell EMC VxRail can offer organizations dramatic benefits and results. In Amarillo, the city cut first-responder response time by 60% and saved $800,000 in hardware-refresh costs for its fleet of police cars. It also improved productivity by reducing PC boot-up times from 10 minutes to 36 seconds.

VxRail also helped when the pandemic created a sudden need for some essential employees to work remotely.

“Of course, large numbers of employees remained essential workers, but for those who could work from home, we were able to stand them up practically overnight, which would’ve been impossible before our IT modernization,” Gagnon says. “In fact, we spun up a 75-person virtual call center for remote public-health clinics in just a day and a half using our VxRail-supported VDI.”

Adopting a modern data management solution like Dell EMC VxRail can offer organizations dramatic benefits and results.

Proven Systems Offer Organization-Wide Benefits

Build-your-own HCI systems can be complex. By choosing a pre-built, pre-tested solution, organizations can quickly adopt a reliable infrastructure for business-driven operations (Video 1).

Video 1. A combination of agility, operational consistency, and support helps organizations deploy an HCI system quickly. (Source: Dell Technologies)

VxRail is ready for application development for DevOps or IT teams. In the commercial space, it’s ready for databases and mission-critical activities, like SQL, Oracle, SAP HANA, Exchange, and SharePoint. And in enterprise, it’s used for those applications as well as edge, AI, and general-purpose workloads.

When looking for a solution, deployment and management are often top concerns for organizations. “Operational consistency is important,” says Justin Jones, VxRail marketing team leader for Dell Technologies. “VxRail is the only jointly engineered HCI system built to enhance VMware. As the leader in the virtualization market, most customers are already using VMware in-house. For those users, VxRail delivers a familiar operating experience.”

Making decisions with the future in mind can allow organizations to focus on growth without worrying about scale. VxRail is the first hyperconverged system fully integrated with VMware Cloud Foundation SDDC Manager, and it delivers a direct path to the hybrid cloud from an automated platform.

“A lot of customers know they need to grow on-premises, but they also want the ability to grow in the public cloud,” says Jones. “VxRail supports their applications, their workloads, and ultimately their business with a hybrid cloud.”

And for organizations that require holistic support, choosing a solution from a reputable provider like Dell Technologies can offer advantages, such as peace of mind.

“System support from hardware to hypervisor to hybrid cloud is a hallmark of VxRail, and 97% of problem cases are resolved in-house,” says Jones. “With 25,000 hours of testing and hundreds of engineers validating each release, your IT department isn’t going to be stuck having to read white papers or docs to determine how to run, upgrade, or resolve issues. This moves them away from siloed teams doing separate research so they can focus on what they do best—delivering business value.”

HCI Leads to a Data-Driven City

The transition to an HCI architecture is a game changer for organizations like the City of Amarillo. From improving public safety to quick pandemic response, a modern IT infrastructure doesn’t just lower operational costs, it enhances the lives of citizens. And the city is looking to the future.

“Our next big moves are around expanding automation, both for operations and security, and our use of analytics,” says Gagnon. “We want to be a data-driven city, and our VxRail modernization is helping to get us there.”

Digital Transformation for SIs: Think Big, Start Small

Savio Lee Systems Integrator

[Podcast Player]

Turning digital transformation initiatives into new business is the North Star for many systems integrators. What if you could deliver holistic end-to-end solutions—from proof of concept to final deployment—without huge investments?

In this podcast we explore the possibilities with Ingram Micro, a global IoT solutions aggregator supporting SIs, ISVs, and OEMs. In our conversation with Savio Lee, Senior Business Leader for Digital Transformation, we explore:

  • How to accelerate digital transformation by taking an architectural approach
  • Where SIs can fast-track their IoT practice with minimal investments
  • Why thinking big and starting small delivers customer ROI more quickly
Apple Podcasts  Spotify  Google Podcasts  

Transcript

Savio Lee: But at the end of the day we encourage our customers to think big, but allow them to start small while mitigating the risks.

Kenton Williston: That was Savio Lee, Senior Business Leader for Digital Transformation at Ingram Micro Canada, and I’m Kenton Williston, the Editor-in-Chief of insight.tech. Every episode on the IoT chat I talk to industry experts about the technology and business trends that matter for developers, systems integrators and end users. Today I’m talking to Savio about the way systems integrators can help their customers accelerate digital transformation by taking an architectural approach to the Internet of Things, and what this means in terms of new business opportunities. We’re also going to dig in to the ways you can rapidly create proofs of concept without blowing your budget in applications like industrial automation, fleet management, and environmental monitoring. So, Savio, welcome to the show.

Savio Lee: Thank you very much, Kenton.

Kenton Williston: So, like your title, there’s quite a lot going on there—Senior Business Leader for Digital Transformation. What exactly does that mean, and how does that relate to what Ingram Micro does as a business?

Savio Lee: So, as Senior Business Leader for Digital Transformation, I’m currently responsible for leading a business unit focusing on emerging technology, and by emerging technology I’m talking about IoT, mixed reality, robotic process automations, and so on.

Kenton Williston: Oh, that’s very cool. And digital transformation, of course, has been a big buzzword for the last few years, I’d say. I think 2020 really brought it to the forefront, kind of out of necessity—everyone had to rethink a lot of their business models. So why don’t you tell me a little bit about what you see happening in terms of enterprises moving towards digital transformation, and how that relates to the business that systems integrators are in.

Savio Lee: Yeah, I would say digital transformation—the term itself is a very broad term. Organizations are starting to recognize what digital transformation potentially means to them. And they’re able to cut through all the noise and floats around things like IoT and digital transformation, and so on. And there’s time to make investment, and these investments are not necessarily in the millions of dollars to do massive organizational-wide overhaul to help them to become smart overnight. But these are investments in the form of new roles being created—such as chief innovation officer. Or the accounts they’re assigned to put together what is called—commonly called—the innovation team, right? That comprises individuals from different departments and different parts of the business who come together and innovate, right?

And the best thing about these innovation individuals or teams is they are not bound by the typical corporate bureaucracies. And they have their own budgets to conduct various POCs, allowing them to be flexible and move fast. And if they don’t succeed, they move on to the next task. And what this means for the traditional systems integrator is with digital innovation you’re talking about new people. So it’s very important that they understand the macro as well as the social norms, as well as the organizational structure within an organization—to make sure that they start to build good relationships, have meaning around digital transformations—what digital transformation means for the customers, so that the SIs can align themselves accordingly.

Kenton Williston: And what is Ingram Micro doing to help systems integrators through that journey?

Savio Lee: Yep. So at Ingram Micro we play the critical role of an IoT aggregator, right? If you look at IoT as a whole, it creates new opportunities, but with new opportunities also come new complexities, right? Because we are talking about not just new technology, but we’re also talking about new entrance to the market, as well as new stakeholders. So Ingram Micro plays the critical role of an IoT aggregator and an orchestration platform. From an aggregation point of view, it just aligns very well with our core function as a distributor, where we function as a one-stop shop for all the IoT needs from an ecosystem perspective.

As of today, we have an extremely robust IoT ecosystem that spans all the way from the sensors to the cloud and IoT application, and anything and everything that is in between. Now, we have gone a step further in recognizing that, because IoT is also a collection of similar systems, we function as an orchestration platform to unify and pre-integrate the ecosystem. And by doing so we are doing a lot of the heavy lifting, so that by the time it gets to our partners they can focus more on solutions and selling business outcomes.

Kenton Williston: Yeah. So I think Ingram Micro is a pretty well-known name in the marketplace as a distributor. It’s certainly a company that I think of traditionally if I’m thinking of—okay, I need to go get a certain set of parts to put together some kind of solution. But I think what I’m hearing you say is that your business has really moved beyond being a parts catalog, to now offering already-integrated solution stacks, as well as services to support them. Do I have the right idea there?

Savio Lee: Yeah, that’s exactly it. When I started this business unit initially, we recognized very on and very quickly that the traditional fulfillment for just a pick-pack-ship model—that’s not going to run. What the market demands is more around solutioning, right? So, that was sort of a 180-degree turn from what we’re typically used to. And then we started offering out-of-the-box solutions, which is an end-to-end solution that is plug and play from the single vendors. Now, not all vendors can do everything, and where there is a gap, we leverage our ecosystem to bring pre-integrated or integrated solutions to the market as well.

Kenton Williston: Yeah. So I want to talk about this word “solution,” right? This is kind of a catchall term that can mean just about anything from—hey, I’ve got a chip—to something really huge and comprehensive. And I think what’s interesting about what I’m seeing Ingram Micro doing is that you’re really thinking about the approach to the Internet of Things from an architectural perspective—by which I mean, like you were saying, everything from the sensors, to the network connectivity, to the data processing, to whatever else there might be. So, can you give me a sense from your perspective of what it means to have an architectural approach to IoT applications?

Savio Lee: Absolutely. I would say—and the reason we had to take an architectural approach is—as of today, or even a couple of years ago, you would ask 10 different vendors what IoT is, you would get 10 different answers, right? You ask a network vendor what IoT is, they’re going to tell you it’s all about the PoE switches. You ask a cloud company what IoT is, they’re going to tell you it’s all about a graph, right? I am not saying they are wrong. They are right. But the problem is each of them is approaching IoT from their own lens, and defining IoT in their own terms, right? And because of that, it created a lot of confusion in the market in terms of knowledge and skill sets. What we saw is that we really need to take an architectural approach to IoT. And what that means is, not looking at IoT from a particular technology stack; architectural means holistic, end-to-end, flexible, right? And also comprehensive.

Kenton Williston: Yeah. Absolutely. So, can you give me an example of what that looks like in practice from one of the markets you’re serving? Like, say, industrial automation, or whatever you would like to use as an exciting example there.

Savio Lee: Yeah. So I would say in the area of industrial automation, by taking an architectural approach, we start with trying to define what the challenges and objectives of the end customers are like, right? And then from there we try to understand—okay, what are the current processes? What is the current system that’s in place, right? Because IoT is not always about rip and replace. A big part of IoT to achieve ROI is about working with the existing system. By taking an architectural approach we have the entire view of—okay, what is required from a technology point of view that is—. What is required from a technology stack to achieve that business outcome? And what is currently in place? Where are the gaps, and what can we do to fill those gaps—but, most importantly, connect all the dots together.

Kenton Williston: Got it. So it sounds to me like a big part of this is that, if I’m thinking about this from the perspective of a systems integrator, I might have expertise in certain applications. I might have expertise in certain technologies. And what I can get out of working with Ingram Micro is sort of a partner in the space who can help me fill out all the rest of it, so that I don’t have to learn it all myself. I’ve got someone else who’s going to support me through the process of thinking about this on a more, sort of, global scale, if you will.

Savio Lee: Absolutely. You couldn’t be more right on that note. We have all types of partners. We have partners that want end-to-end solutions from Ingram Micro that may be out-of-the-box, or it could be an integrated solution that is built by an Ingram Micro that involves multiple vendors. We also have partners that are specialized in a certain level of the IoT technology stack, but they don’t have the complete picture. right? Those kinds of partners, say as an—. Let’s take an example of an ISV, right? They may have the IoT applications that make sense of all the data collected, but they have challenges in terms of instrumenting the physical world and how to get data to their applications, right?

In that sense, we allow them to be the data or the application of IoT, right? By working with traditional IT vendors—as another example—they want to focus on what they’re best at, which is network and security infrastructure. In that sense, we let them be the IT of IoT. So our approach in terms of partnership—we are very flexible, and we recognize that there’s not really one-size-fits-all.

Kenton Williston: Got it. That makes sense. So, again, I want to kind of dig down, and sort out the practical terms of how this plays out. One of the things that I think is really interesting about what digital transformation really means that you were talking about, is this idea of very quickly trying ideas out and moving on if they don’t work, right? Doing a lot of proofs of concept to see if a certain idea is even going to be feasible. And that certainly makes sense from the perspective of the end customer, whoever it is that’s wanting to explore these new ideas. But if I’m a systems integrator, honestly, that sounds a little frightening, because it’s like—okay, I’m going to get stuff thrown at me left and right. And my customers are going to want my assistance putting together these proofs of concept. How am I going to be able to very rapidly respond to those demands? What do you think about that?

Savio Lee: Yep. We’ve seen that time and time again, and that’s a very good example that you just gave me. And there’s a couple of other examples as well, where customers are looking for IoT solutions but they don’t want a really permanently installed IoT solution, right? So what we have come up with is a program of what is called IoT on Demand, right? So, IoT on Demand is essentially a pre-built, custom-selected set of hardware for different use cases and verticals that customers, or our partners, can partner or leverage to allow them to conduct POC rapidly, right? And at the same time, if they to choose to buy it because the way it’s designed is—because it’s designed to be flexible, it allows the partners to easily repurpose and redeploy the solution and the infrastructure for different use cases.

Kenton Williston: And what about the financial elements of this? Because in addition to the time crunch that I think this would put on systems integrators, it also strikes me that this could be a pretty big financial burden—potentially to have to keep spending a lot of money on spinning up these proofs of concept. Are you doing anything particular to address that end of the challenge?

Savio Lee: Yeah. So, as part of our POC on Demand, which is—part of our on-demand program is we also have what is called a POC on Demand program, where you can turn any budget that you have for conducting POC into an OpEx model. That takes away the need for having to own or permanently buy IoT sensors, and so on, right? We have a large pool of IoT devices, and so there’s technology for different use cases—pre-built solutions that allow our systems integrators to leverage and conduct POC rapidly without making significant investment. And when the POC is done, they simply have to return the equipment and everything to Ingram Micro.

Kenton Williston: Got it. And presumably this is kind of like a win-win-win scenario. The end customer gets to explore all of these ideas very quickly. The systems integrator gets to assist them and help them arrive at a final architecture that they want to deploy—we know—without making a huge investment. And then, of course, the systems integrator in Ingram Micro itself gets the benefit of retaining that business over time. And once that proof of concept becomes a real, scalable, deployable sort of idea, then there’s that revenue stream available.

Savio Lee: Exactly. So the idea of a POC-as-a-service is to—by taking the need for owning hardware on a permanent basis—what that essentially does is it allows and frees up the budget to conduct multiple POCs simultaneously, as opposed to just one POC simply because they have to buy the equipment.

Kenton Williston: Yeah, totally. And I think I just see pretty much everywhere, folks moving to as-a-service kinds of offerings, and I think it totally makes sense to offer that here in the IoT space as well. Now, something that I’m really curious about—we’re talking about having an architectural-level viewpoint of things, and on the hardware side that kind of makes sense to me how you would do that—on the, like, say, cloud platforms and the software platforms point of view—that makes sense to me. But there’s all this glue that needs to be put in between to make all these parts work together and to surface actionable information to the end customer. So, is there anything that IoT systems integrators can do to make that sort of last step in the process of bringing everything together simpler?

Savio Lee: Yes. Absolutely. That’s actually a very good question, because one of the key requirements to build an IoT practice as a solutions integrator is that you must have software development resources or skill sets in-house as well, right? Which the traditional IT systems integrator does not possess, right? So as part of our architectural approach to IoT, we focus heavily on data because we understand the value of IoT is in the data. And in order to mitigate the need to have software development in-house, we launched what is called Project Last Mile, right?

Project Last Mile, which has been completed, was actually a two-year project in Europe. Project Last Mile essentially allows end customers, or our systems integrators, to easily and quickly build custom IoT applications on our IoT platform with little-to-no software coding required. And because our IoT software platform is microservices-driven and it’s modular in nature, we have gone a step further to pre-build templates for different use cases and verticals—different types of IoT solutions that our partners or systems integrators can simply download and customize the last 10%.

Kenton Williston: That’s really cool. So, I assume what you’ve done here is specifically set up these dashboards to work with the rest of the offerings you have. So, for example, if I’m doing some environmental monitoring, let’s say, that you’ve got dashboards that are already able to interface with those environmental conditions sensors. Is that right?

Savio Lee: Exactly right. We would have, for example, a pre-built template for fleet management or asset tracking, right? Because the reason we have to pre-build some of these templates is because, as important as, and as cool as IoT is, the other missing part of the equation is domain expertise, right? By taking this templated approach, we incorporate some of the best practices into the template in terms of how the data should be represented, and so on.

Kenton Williston: Yeah. So that’s a really good point. I think that’s one of the key value-adds that a systems integrator will bring to the end customer—having a lot of domain expertise. But I think one of the challenges with digital transformation is that it brings together such a broad scope of what that end customer might be doing, that a systems integrator may have expertise in one area—so it’s like, take the fleet management, I think it’s a good example—maybe you’ve got a really good expertise in, say, onboard monitoring of the fleet to make sure the engine is running fine, the drivers not driving too fast, etc.

But now the end customer wants to also track the location, and more efficiently route all their vehicles. Well, maybe you don’t have that expertise. So it sounds like what you’re saying is—don’t worry about it. We’ve thought about these problems already, and we can help you not only just deploy these systems with all the different bits and pieces where you may or may not have expertise in the specific technologies, but we’re also going to bring some domain expertise to the table.

Savio Lee: Exactly. And on the topic of domain expertise, we have a separate program, what is called the IoT Co-Creation program. The IoT Co-Creation program is really our strategy of how we bring in domain expertise to build various applications, and help our systems integrators and their end customers derive additional value from the data. Take the example of energy management, right? I mean, we are great at visualizing all the different types of data coming in from the sensors, but how to represent that in the context of energy management—that’s a whole different story, right? For example, how do you convert gas into carbon consumption? What is acceptable in that domain? What is bad? What is the threshold, right? And that’s really where the IoT Co-Creation program comes in—to help us to co-create programs, and allow us to represent data in proper context.

Kenton Williston: Yeah. That makes sense. Because I think—and when you get down to the fundamentals of what digital transformation is about, it’s partly about the digital part: digitizing things, gathering that data. But then you’re not really transforming much if you’re just gathering data—you need to have it in a form that’s actionable and intelligible so that you can actually do something with it.

Savio Lee: You’re absolutely right.

Kenton Williston: So I love this example about the energy management. I know there’s a whole bunch of other markets that you’re very active in having some pretty sophisticated platforms and templates already built—like fleet management that I talked about. Any others that you think would be really good examples to highlight here?

Savio Lee: Yeah. A really good example where we get a lot of traction is in the area of AI-based video analytics, right? I may be getting a lot of surprise here, but at the end of the day, if you look at the IP cameras, it’s really an optical sensor, right? The reason we are able to get that, we’re able to get a lot of traction and interest from customers—that includes cities and organizations as businesses—is because the smart video analytics leverage the existing IP infrastructure, right? And similar to how the single pair of human eyes can perform multiple functions, it’s exactly the same thing with the Intel-based video analytics as well. The project that we have done that leverages Intel architecture, AI-based video analytics in the area of smart cities would be, for example, smart traffic counting. So we actually use regular IP cameras that are mounted on a pole.

We fit a live video stream into an AI engine that has actually analyzed the video and recognized and counted objects such as pedestrians, cyclists, vehicle types, direction of the objects. The data that’s generated by the AI engine is pushed to a dashboard for further slicing and dicing of data, and visualization that allows the cities to use that data to help them optimize their urban planning. If there are a lot of cyclists in here—maybe if they see a lot of cyclists riding on the roads or on a pedestrian pathway or sidewalk, maybe there’s a need for a bike lane, right? They can also use those data to help them optimize the maintenance schedule for those roads that have cyclist bike lanes, right? Whether they need to have any maintenance scheduled, especially during winter to make sure there is no snow in those bike lanes.

Kenton Williston: Got it. That totally makes sense. So, something I’d like to do now is dig just a little bit deeper. We’ve been kind of glossing over, I think, what it means to have an architectural approach. And I’d like to kind of dig a layer deeper and talk about some of the specific elements of the technologies and expertise that need to go into overall architecture, and just highlight some of the places where I think Ingram Micro has some interesting things going on. So, first of all, fundamentally, if you’re going to do anything in the IoT, it’s all about taking real-world data and digitizing it.

So of course there’s sensors needed, and sort of those Edge processing elements needed, to take that raw data and turn it into something consumable. And of course, as we’re talking about, that’s really Ingram Micro’s history and fundamental strength, and you’ve got that huge catalog of parts. So that kind of checks that box for me. But then the next thing—and you talked about this early on with this idea of orchestration. So I’m wondering if you could talk to me about, what does that mean? And how does that relate to deploying IoT systems?

Savio Lee: So, by taking an architectural approach to IT, it means, as I mentioned earlier, an end-to-end approach—unified, holistic, and flexible, right? And diving a lot deeper than what I just mentioned, leveraging our strength as a distributor, we innovated—. We looked at IoT as a whole, identified the areas where there are challenges that Ingram Micro could bring value to the table, right? The number one challenge would be instrumentation of the physical work. As much as 70% of the IoT projects fail simply because you cannot get data—not the application problem, it’s the data. So we have 200-plus ready-to-deploy sensors and gateways, right? For all types of use cases and verticals. That challenge is also about reliable hardware and quality data.

So we all make sure that we use tested, industry-certified and ready-to-deploy standards, right? Now, the next piece is, as you know, with great complexity also comes lack of industry standards, and so on. So what we have is really a middleware, or an orchestration platform, which connects our ecosystem of hardware vendors together, right? And that itself allows us to address the challenges of lacking technology standard in terms of protocols, and allows us to normalize, to unify the data and send it to any applications or to our own applications.

Kenton Williston: Got it. That totally makes sense, and I agree, that 70% failure rate surprises me—how big of a number it is. But I know that that is indeed one of the biggest challenges for folks—is just being able to gather the data and get the various systems that are gathering that data to even talk to each other in the first place. Major, major challenge.

Savio Lee: Yeah. Absolutely. And we’ve gone multiple steps beyond that, right? I mean, it’s great we have an orchestration platform that serves as an abstraction layer for our ecosystem of partners, because the number one rule we have for all our technology vendors is that they have to have an open ecosystem approach to share the data, right? So that’s really our orchestration platform that unifies the ecosystem to create one common data layer. The next thing is we also have what is called a plug-and-play approach that allows end customers and systems integrators to easily onboard IoT devices by simply scanning the QR code, right? So no longer do you have to go and upload and find a serial number and type in that 16—alphanumeric device ID, which can be time consuming and error prone as well.

Kenton Williston: Yeah. And I think that’s a really good point, and I think it really highlights one of the advantages of working with an aggregator like Ingram Micro. So, I know there are individual technology providers who have these sorts of plug-and-play onboarding capabilities, but obviously it’s only going to be for their own solutions. And what you can offer as an aggregator is the ability to do this for all the solutions that are in your portfolio.

Savio Lee: That’s exactly it, right. It really goes back to your comment that we serve as an orchestration platform—obviously aggregating the ecosystem aligns very well to our core function as a distributor. But as an orchestration platform we sort of pull all the pieces together and unify them.

Kenton Williston: Yeah. And the other thing I wanted to ask about: we’re talking about the challenge of getting elements at the Edge to talk to one another—it’s not just the protocols and the data normalization and things like this that are challenging, but it’s even the networking itself. So what are you doing to help systems integrators deploy networks in these use cases?

Savio Lee: Yeah. So, as you know, there are certain wireless networks or non-wireless networks that are very specific to IoT. So, what Ingram Micro is doing is that we are actually standing up the infrastructure to provide a wireless network for our systems integrators as well as the end customers, because the network is a very critical component of the entire IoT infrastructure. And we feel that we have the size, as well as the financial resources, along with the technical resources to take away a lot of the heavy lifting required just to stand up IoT networks. So by taking that burden away from them, we allow our systems integrators and your end customers to quickly tap into a network that is managed by Ingram Micro, and get the data to the cloud fast.

Kenton Williston: So, if I can kind of wrap all of this up. I think a big element of what I’m hearing here is that, whether you’re thinking from the perspective of the end customer or the systems integrator, a lot of the reason that digital transformation efforts fail—a lot of the reasons that proofs of concept fail—isn’t because the fundamental idea is a bad idea, but just because there are basic technological hurdles to get past that are pretty difficult for an end customer to get past, and maybe even challenging for the systems integrator to get past because they touch on areas where they maybe don’t have all the expertise—or even just things like talking about this onboarding process where, they can do it, but it’s just a question of can they do it fast enough and cost-effectively enough?

And what you’re doing is really kind of clearing the field by doing these basics of a technology by creating an OpEx sort of model, so that people don’t have to make massive investments. So that you can actually rapidly innovate, and not just get totally stuck at step one of the basics of technology.

Savio Lee: Absolutely. Right. I mean, we recognize the fact that digital transformation is not an overnight thing, and it means different things to different people. Everyone is talking about smart cities, smart manufacturing, maybe even smart galaxy—whatever, right? But at the end of the day we encourage our customers to think big, but allow them to start small while mitigating the risk. So that for each small modular project that they do, they are realizing ROI and how to optimize it before they go to production.

Kenton Williston: Yeah. That makes sense. And I’m also wondering—it seems like there’s an opportunity here for the systems integrator to rethink their business model, and what kind of offerings—and even the kinds of opportunity space that’s available to them. And it strikes me that there’s probably a piecemeal way of attacking all of those opportunities as well. So I’d love to hear your thoughts on, sort of, the step-by-step approach that systems integrators can take to expand their capabilities—expand the opportunities they can address without biting too much off at one time.

Savio Lee: So, the opportunities presented by IoT are certainly broad. It’s very important that the systems integrator must understand the picture. And, in the grand scheme of things, understand where they want to focus on, what they want to build off, right? The quickest route to market is really understanding that the customer really wants end-to-end solutions. They’re not looking for technology. They’re not looking for sensors. They couldn’t care less about what the sensor looks like, right? And the good thing is, we have pre-built, out-of-the-box solutions that allow you to easily sell to the customers—allow them to try it, along with our various innovative services and programs to help support you all the way from the sales to implementation and post-sales.

Our approach is really learn-as-you-walk, right? Or learn-as-you-grow type of thing, right? Without making any significant investment, because we have that infrastructure in place to help them build. And over time, hopefully our system integrated, they will realize that—once they understand what is required to succeed in IoT, then they can decide what role they want to play in IoT. Do they want to focus on—if it’s an IT systems integrator—do they want to just focus on IT and be the IT of IoT by just focusing on the network and infrastructure for IoT, right?

Or do they want to be a solutions integrator over time by building additional, and hiring additional, resources in-house, such as op software development, and so on, right? And there are going to be guys who say, You know what? We don’t want to be the solutions integrators; we want to focus on the data because that’s ultimately where the value of IoT is. And they want to build a data practice by helping the customer derive additional value from the data, and to drive organizational change in efficiency.

Kenton Williston: So, I’m wondering also how systems integrators can bring this story—put the scope of what they can do by partnering with an aggregator like Ingram—in front of their customers, to help their customers reconceptualize the relationship they have with the systems integrators and the sort of projects they can embark on. Do you have any thoughts on that?

Savio Lee: Yeah. Absolutely. IoT is anything but a fulfillment-for-business for Ingram Micro Canada. As part of a high-touch, white-glove process, we take a very collaborative approach with our systems integrators in terms of engaging with the end customers. What I’m trying to say is, we are actually out there course selling, engaging the end customer directly a lot with our systems integrators. And because we have a seat at the table for the end customer, we get an in-depth look at what the opportunities are, what is required for the end customer to succeed.

And we are able to hand-hold them—architect the entire solution end-to-end, and work with a systems integrator to deploy the solutions. And we’ve seen it time and time again, where this high-touch collaborative approach really works after a couple of times, and all of a sudden the systems integrator gets it—they start to have that business-outcome discussion with the end customers outside of IT. That includes the lines of business as well.

Kenton Williston: Perfect. So this has been really great. I feel like I’ve got a whole new perspective on what digital transformation even is—let alone how systems integrators can approach it. Any questions that you wish I had asked you?

Savio Lee: Yeah. I think one of the key messages I want to get out there is that IoT—to our systems integrators—is that IoT is a journey and not an overnight thing, right? Because we have created that ecosystem, done the job of unifying the ecosystem, innovating at every layer of the IoT technology stack along with the capabilities for our IoT applications. What that means is, partnering with Ingram Micro—we have essentially fast-tracked your IoT practice by a minimum of two to three years, right? And the best part is, you can start your IoT practice with little-to-no investment because of our high-touch, white-glove approach, where we’re actively out there doing a lot of course selling, engaging any IoT opportunities that you may come across so that you are really learning as you grow.

Kenton Williston: Perfect. I love that. With that, I’d just like to say, thank you so much, Savio, for joining us today. This has been really interesting and informative.

Savio Lee: Thank you very much, Kenton.

Kenton Williston: And thanks to our listeners for joining us. To keep up with the latest from Ingram Micro Canada, follow them on Twitter @IngramMicroCA. And if you enjoyed listening, please support us by subscribing and rating us on your favorite podcast app. This has been the IoT Chat Podcast. We’ll be back next time with more ideas from industry leaders at the forefront of IoT design.

Driving Through Efficiencies in Coffee Service

Drive-thru technology is helping businesses improve customer service, reduce waiting times and personalise their marketing, as Costa UK has discovered.

The UK might lag behind the US when it comes to the prevalence of drive-thrus in the hospitality space, but relatively recent entries into this area by coffee giants Costa and Starbucks in the UK as well as growing use of those operated by the likes of McDonald’s and KFC has meant that they are growing in importance.

The pandemic has played its part in this, with McDonald’s becoming reliant on its drive-thrus and click & collect during the third UK lockdown having closed its restaurants for customers even for takeaway.

“When the virus happened that accelerated everything,” says Chloé Bisiaux, head of drive-thru products at Acrelec, which provides drive-thru systems as well as self-ordering kiosks to the food and beverage industry across the world.

“Demand has been rising for drive-thrus as many restaurants have been closed at some point and they have had to find a way to sell their products.”

“Some restaurants had a drive-thru that was not that efficient or lacked investment because most customers would come into the restaurant to place an order. The pandemic has changed that and encouraged companies to look more closely at their drive-thrus. Across Europe there was a big realisation that companies need to invest in the drive-thru.”

What’s next for drive-thrus? @AcrelecUK sees a world of high-tech possibilities. via @insightdottech

As well as making drive-thrus a much more important part of a restaurant group’s offer, the pandemic has also put more pressure on them, and this is where Acrelec, has stepped up to help businesses with products that include outdoor digital menu boards, customer identification technology and systems that bring vital new efficiencies to drive-thru operations.

For example, its digital menu boards are designed to help improve visibility of products, changing their displays based on the time of day. Its boards also have an order view function so that the customer can see their order written down as it is being taken, which helps reduce the number of errors with the order, says Bisiaux. They can also be linked to a restaurant’s POS system so if an item is out of stock that information can be displayed on the screen, so the customer knows not to order it, reducing order time.

To further speed up the drive-thru process Acrelec has created customer identification technology Drive ID, designed to identify customers upon arrival at the order point and generate up-sell opportunities through personalisation.

“Prior to having this type of solution, a customer would have a basic scanner scanning a QR code on a phone if they wanted to redeem an offer and that would very often happen at the payment window,” says Bisiaux. “So, they’ve placed the order already and then realised that they’ve got some chicken nuggets for free – the big problem with that is that it’s not part of the order so [the restaurant] will have to start preparing the nuggets a bit later in the process.”

Drive ID uses Bluetooth-enabled customer identification so when the customer places the order, they can simply select the offer they want to redeem on their phone, which then goes through. Acrelec is in discussion with major QSR brands in the US and Europe on a global level for Drive ID and says it is looking at how it can integrate it with the fast food giant’s mobile app.

AI and Machine Learning

As with self-order kiosks, companies are looking at opportunities to upsell at the drive-thru through tailored product recommendations. Bisiaux says Acrelec is looking to harness AI and machine learning so that the machine would learn what items are purchased together and as a customer places an order they would be given a recommendation made specifically for them. “Not only does this make the process simple but it increases the average check.”

She also says that Drive ID is able to recognise a customer (if previously consented) and, based on their previous orders, simply recommend them their usual order as option, creating a more personalised experience as well as speeding up the order process.

Acrelec’s QTimer 2.0 is another product designed to further streamline the drive-thru experience by helping businesses find out the duration of a customer order at different points – order, payment, and service. It also gives businesses data on the overall length of service, which Bisiaux says helps teams see what’s happening in real time and identifies whether there are any bottlenecks.

“They can go back and look at analytics and see if in the evenings they are slower and find out how to speed up the process by adding more staff, or identify times when they are very efficient.”

Coffee chain Costa is one company that is looking to harness the power of QTimer 2.0. Coffee Lovers, a franchise partner of Costa UK that operates 11 venues under licence, began using the technology at its new Coalville, Leicestershire, store in December and says it has been assisting with its customer throughput.

“When a car arrives there’s tech in the ground outside that picks up that car, then it proceeds to the order screen where have dialogue, then payment and collection and then they drive off,” says Ghanshyam Ramparia, Coffee Lovers director.

“There are only three of four pinch points but Acrelec’s solution visualises that journey and gives us data on the back of this, such as how big the queue is and how long customers are waiting at each pinch point, to make that journey as quick and efficient as possible.”

Ramparia says that QTimer 2.0 tells provides information on the customer journey which then allows staff to act correspondingly in terms of service.

“If there’ a bottleneck and a lot of people who have already placed their order are waiting for their drinks, the person taking orders is directed to take more time with ordering customers and ask them more questions to buy some time,” he says.

“Or if there is a big pile up of cars waiting to order and no one waiting to collect we might want to get customers off the speaker as quickly as possible. QTimer 2.0 visualises that.”

QTimer 2.0 can also be linked to Acrelec’s digital menu boards and content management systems for even greater functionality. “If the drive-thru is busy you can say you want to have a more simplified menu and recommend items that are quicker to prepare and guide customers to make a decision more quickly,” says Bisiaux. “When it is emptier you can advertise the more premium products and have a wider selection.”

Outdoor digital signage plays a major role in enticing customers as well as showcasing exactly what’s available during the day part including the exclusion of any items that is unavailable.

Simplicity is key to keeping customers moving as well as making product selection easy, according to Farhaan Mohideen, product strategy and IoT spokesperson at Intel. “Intel provides the right hardware form factor and integrated graphics to enable media processing much richer supporting different media formats on digital signage,” he says.

Intelligent Integration​

A drive-thru requires many different systems to be exchanging data to enhance the customer experience and Acrelec has developed an advanced solution that stitches together information from multiple sources and make decisions in near real time.

“We wanted something that was integrated, reliable and flexible,” says Ramparia. “Acrelec has got a lot of technologies that can talk to other technologies; rather than have four or five vendors trying to work together we would rather have it all under one umbrella where it’s proven that it all works together.”

Mohideen says that Intel’s technology plays a part in integration. “Intel has been advocating the use of Edge compute to drive workload consolidation and low latency edge inference that is able to act on the data produced by the various IoT devices and sensors.

“Designing a drive-thru system requires much consideration on not only what’s happening as the customer drives in but capacity in the kitchen and availability of produce for meal preparation. Intel’s compute, hardware acceleration, Intel® Distribution of OpenVINO toolkit and Intel vPro® technology are key foundational elements in modelling and consolidating the various workloads and eliciting key insights at the drive-thru and the kitchen.”

Such insights are what Ramparia hopes to take advantage of at the Coalville Costa. “With an integrated solution we can react to day parts, the weather, the time of day, and customise offers and promotions on the back of it. If it’s a sunny day we can promote cooler drinks, for example, using AI to customise the offer.”

Ramparia also hopes to take advantage of technology that will enable number plate recognition or use Bluetooth to recognise returning customers and even to create local offers based on how much stock a particular site is holding. If there’s a glut of cakes, for example, they can be automatically promoted to customers as an add-on.

“Using technology more creatively in the future we know that there’s a lot more that can be done. A drive thru serving coffee seems a simple thing but there’s a lot more that goes on behind the scenes.”