Tag Archives: robotic

RPA in manufacturing increases efficiency, reduces costs

Robotic process automation software is increasingly being adopted by organizations to improve processes and make operations more efficient.

In manufacturing, the use cases for RPA range from reducing errors in payroll processes to eliminating unneeded processes before undergoing a major ERP system upgrade.

In this Q&A, Shibaji Das, global director of finance and accounting and supply chain management for UiPath, discusses the role of RPA in manufacturing ERP systems and how it can help improve efficiency in organizations.

UiPath, based in New York, got its start in 2005 as DeskOver, which made automation scripts. In 2012, the company relaunched as UiPath and shifted its focus to RPA. UiPath markets applications that enable organizations to examine processes and create bots, or software robots that automate repetitive, rules-based manufacturing and business processes. RPA bots are usually infused with AI or machine learning so that they can take on more complex tasks and learn as they encounter more processes.

What is RPA and how does it relate to ERP systems?

Shibaji Das: When you’re designing a city, you put down the freeways. Implementing RPA is a little like putting down those freeways with major traffic flowing through, with RPA as the last mile automation. Let’s say you’re operating on an ERP, but when you extract information from the ERP, you still do it manually to export to Excel or via email. A platform like [UiPath] forms a glue between ERP systems and automates repetitive rules-based stuff. On top of that, we have AI, which gives brains to the robot and helps it understand documents, processes and process-mining elements.

Why is RPA important for the manufacturing industry?

Shibaji Das, UiPathShibaji Das

Das: When you look at the manufacturing industry, the challenges are always the cost pressure of having lower margins or the resources to get innovation funds to focus on the next-generation of products. Core manufacturing is already largely automated with physical robots; for example, the automotive industry where robots are on the assembly lines. The question is how can RPA enable the supporting functions of manufacturing to work more efficiently? For example, how can RPA enable upstream processes like demand planning, sourcing and procurement? Then for downstream processes when the product is ready, how do you look at the distribution channel, warehouse management and the logistics? Those are the two big areas where RPA plus AI play an important role.

What are some steps companies need to take when implementing RPA for manufacturing?

Das: Initially, there will be a process mining element or process understanding element, because you don’t want to automate bad processes. That’s why having a thought process around making the processes efficient first is critical for any bigger digital transformation. Once that happens, and you have more efficient processes running, which will integrate with multiple ERP systems or other legacy systems, you could go to task automation. 

What are some of the ways that implementing RPA will affect jobs in manufacturing? Will it lead to job losses if more processes are automated?

Das: Will there be a change in jobs as we know them? Yes, but at the same time, there’s a very positive element that will create a net positive impact from a jobs perspective, experience perspective, cost, and the overall quality of life perspective. For example, the moment computers came in, someone’s job was to push hundreds of piles of paper, but now, because of computing, they don’t have to do that. Does that mean there was a negative effect? Probably not, in the long run. So, it’s important to understand that RPA — and RPA that’s done in collaboration with AI — will have a positive impact on the job market in the next five to 10 years.

Can RPA help improve operations by eliminating mistakes that are common in manual processes?

Das: Robots do not make mistakes unless you code it wrong at the beginning, and that’s why governance is so important. Robots are trained to do certain things and will do them correctly every time — 100%, 24/7 — without needing coffee breaks.

What are some of the biggest benefits of RPA in manufacturing?

Das: From an ROI perspective, one benefit of RPA is the cost element because it increases productivity. Second is revenue; for example, at UiPath, we are using our own robots to manage our cash application process, which has impacted revenue collection [positively]. Third is around speed, because what an individual can do, a robot can do much faster. However, this depends on the system, as a robot will only operate as fast as the mother system of the ERP system works — with accuracy, of course. Last, but not least, the most important part is experience. RPA plus AI will enhance the experience of your employees, of your customers and vendors. This is because the way you do business becomes easier, more user-friendly and much more nimble as you get rid of the most common frustrations that keep coming up, like a vendor not getting paid.

What’s the future of RPA and AI in organizations?

Das: The vision of Daniel Dines [UiPath’s co-founder and CEO] is to have one robot for every individual. It’s similar to every individual having access to Excel or Word. We know the benefits of the usage of Excel or Word, but RPA access is still a little technical and there’s a bit of coding involved. But UiPath is focused on making this as code free as possible. If you can draw a flowchart and define a process clearly through click level, our process mining tool can observe it and create an automation for you without any code. For example, I have four credit cards, and every month, I review it and click the statement for whatever the amount is and pay it. I have a bot now that goes in at the 15th of the month and logs into the accounts and clicks through the process. This is just a simple example of how practical RPA could become.

Go to Original Article
Author:

Promises for 5G in healthcare are great, but it’s still early days

Imagine a patient lying on an operating table, a robotic arm hovering above, transmitting real-time video to a doctor who is 300 miles away performing the operation.

Although it’s years away from reality, remote robotic surgeries like this may be possible, thanks to the high-speed, low-latency capabilities of the fifth-generation cellular wireless network, according to Chris Antlitz, a telecom analyst at consultancy Technology Business Research Inc.

“There are all kinds of different use cases for 5G in healthcare, some of them are a little more science fiction-y than others,” he said.

As with the arrival of previous generations of wireless networks, carriers like Verizon Wireless and AT&T are touting the benefits of a 5G network, including connecting more devices to the bandwidth, faster real-time connections and less buffering. The potential seems great, but it’s still early days. The network is only just being rolled out.

Analysts believe it will be at least three years before the 5G network gains traction and more widespread commercial availability, but even longer before it takes hold in healthcare. While healthcare CIOs don’t need to decide whether to invest now, experts say it’s not too soon to start planning and talking to carriers.

What is 5G?

There are all kinds of different use cases for 5G in healthcare, some of them are a little more science fiction-y than others.
Chris AntlitzTelecom analyst, Technology Business Research

5G is an upgrade to the current fourth-generation (4G) Long-Term Evolution (LTE) wireless network.

The 5G network will transmit data at a theoretical peak of 20 GBps, while 4G only reaches a peak speed of 1 GBps. Latency, or the delay before data transfer, is also expected to drop from 10 milliseconds with 4G to 1 millisecond with 5G. The new network will offer more available bandwidth, which means an uptick in the amount of data transmitted.

The 5G network will rely on small cells, or wireless transmitters and receivers, to transmit large amounts of data at high speeds over short distances. According to Verizon’s website, the carrier has spent the last several years building out small cell sites in towns and cities to enhance the current 4G LTE network, while also laying the groundwork for 5G. While cell towers provide coverage over long distances, small cells enable stronger coverage in more densely populated areas.

Small cell technology supports transmitting signals over millimeter wave spectrum, a high-frequency radio wave with a short range that the 5G network will use to transmit more data at faster speeds. Small cells are roughly the size of a mini fridge, according to Verizon, and are mounted on tall structures, like utility poles and rooftops.

“It’s going to be faster,” Antlitz said. “Your download times are going to get lower. There’s going to be less latency, so the buffering is going to be practically nonexistent when you’re downloading files or streaming.”

Rajesh Ghai, research director for IDC’s carrier network infrastructure research program, said 5G’s abilities to connect more devices and people together will affect some areas of healthcare, such as virtual reality. Yet these new capabilities will require 5G network slicing, which Ghai said won’t occur for the next two to three years.

Network slicing enables a carrier to offer customers a virtual network or a portion of the network to attain, in the case of 5G, greater speed and data capacity. Doing so would mean a healthcare organization could continue to be a 4G LTE shop but use 5G for specific use cases, he said.

5G could benefit existing capabilities in healthcare

The 5G network will enhance existing capabilities in healthcare, and Technology Business Research’s Antlitz believes CIOs should look to telemedicine as a “low-hanging fruit” and initial 5G use case.

“5G can revolutionize healthcare from that perspective,” Antlitz said. “It’s basically just video conferencing. It’s a real-time, high-resolution, no buffering type experience that you can’t get with 4G.”

5G features and benefits for healthcare

Jeff Becker, an analyst at Forrester Research, echoed Antlitz’s comments. He said telehealth is limited by low speeds and internet connectivity issues, especially in rural areas.

“We would see 5G as an opportunity to alleviate that problem,” he said. “Expanding access to high-speed internet to support telehealth is one area where you’ll see folks starting to get excited.”

Becker said another area of impact for 5G in healthcare is remote patient monitoring, which is closely coupled with telehealth.

Remote monitoring tools are already used today, but 5G will expand monitoring capabilities, Becker said. Today, patients with diabetes log their blood sugar levels over time and then bring the logs to their next appointment. Next-generation monitoring tools will feed data in real time into a healthcare organization’s cloud portal where it’s analyzed by AI algorithms. But because of the volume of data generated, the tools require more bandwidth, something 5G brings to the table, he said.

“In remote areas where we’re limited on how well we can stand up remote patient monitoring programs, there’s another area for opportunity,” he said.

Antlitz believes cutting the wire on medical devices such as X-rays and MRIs within a hospital environment is another potential use case, saying that it “opens up all kinds of different scenarios in terms of why 5G could be superior to how we’ve been processing and transmitting data that comes in from this equipment.”

According to a letter from the College of Healthcare Information Management Executives (CHIME), making medical diagnostic equipment mobile so that they can be taken to patients’ rooms will result in productivity gains. The letter, sent to a Senate subcommittee in February, detailed where CHIME officials foresee the greatest impacts of 5G in healthcare.

Making the impossible, possible

While telemedicine and remote patient monitoring may provide some of the first concrete use cases for 5G in healthcare, the new network is also being tied to more far-fetched capabilities as well, such as remote surgery.

IDC’s Ghai said doctors will potentially be able to operate on patients from remote locations, thanks to 5G’s promises of reduced latency, as well as instant high-definition image transmission.

Although more of a skeptic, Forrester’s Becker said the advancement in remote surgery could play a significant role in war zones or natural disaster areas where the hardwire infrastructure has been compromised.

Frost & Sullivan analyst Michael Jude was even more skeptical. He said healthcare organizations would be “crazy” to try remote surgery over a wireless network. Connection reliability rather than latency is a bigger issue, he said.

Although carriers say 5G will be more reliable than 4G, Jude said the required antenna density to perform something like a remote surgery would be “extraordinary,” and antennas would have to be built on “practically everything.”

“I would characterize telesurgery by wireless as being sometime in the not near future because I think it would be really hard to do with any kind of acceptable reliability,” Jude said.

Remote surgery isn’t the only sci-fi example of using 5G in healthcare. Layering artificial intelligence and virtual reality with a 5G network could further enable haptic technology, or artificial touch technology, Antlitz said.

If a small camera inserted into a patient’s brain transmits a 5G signal, AI algorithms can create a three-dimensional model of the brain in virtual reality, helping the doctor better identify anomalies such as tumors. Traditionally, doctors look at a two-dimensional screen, Antlitz said.

“With two-dimensional, you lose some of the depth perception and you can’t see certain things, like around corners, and the veins, their trajectories within that environment,” he said. “So it could improve patient outcomes dramatically.”

And yet despite the promises of 5G in healthcare and despite the new network’s potential impact on the industry, analysts caution CIOs: The road ahead is still long.

Go to Original Article
Author:

HR chatbots from Google, IBM to be in the spotlight at HR Tech 2018

The role of big vendors, such as Google and IBM, in HR technology is expanding as their expertise in conversational robotic intelligence powers some of the chatbots used in HR applications. That observation will be evident this week at the HR Technology Conference & Expo in Las Vegas where HR chatbots will be in the spotlight.

The tech giants’ relationship to HR chatbots is analogous to Intel’s role with PC makers that slap “Intel Inside” stickers on their laptops. The machine learning and natural language processing (NLP) technologies developed by large technology sellers give chatbots conversational capabilities.

“A chatbot stands and falls with the quality of the dialogue,” said Holger Mueller, principal analyst at Constellation Research. “Users will drop and not use [a chatbot] if the answers don’t make sense,” he said.

Conference attendees assessing HR chatbots, in effect, make two bets on any one application. They not only evaluate the HR application but also the capabilities of the vendor that built the underlying, AI-related chatbot technology, whether it’s from Amazon, Microsoft, IBM, Google or some other provider. This technology is key “for the whole solution to work,” Mueller said.

Google’s new Dialogflow powers conversational recruiting

A chatbot stands and falls with the quality of the dialogue.
Holger MuellerPrincipal analyst, Constellation Research

Earlier this year, Google, for instance, announced general availability of its Dialogflow Enterprise Edition. This is Google’s platform for creating voice and text conversation and is based on its machine learning and NLP development.

Google’s technology was adopted by Brazen Technologies, which provides online hiring chat events and a recruiting platform. In late August, Brazen announced a “conversational recruiting” capability based on Google’s system, which provides the underlying chatbot intelligence.

The chatbot conversational capability is assisted by human recruiters who prewrite answers to expected questions that a candidate might ask. The system also conducts an initial screening to try to find qualified people, said Joe Matar, director of marketing at Brazen. He expects the capabilities of conversational HR chatbots to improve rapidly, but it will be a long time before they replace a recruiter’s core skills, such as relationship building, he said.

IBM Watson powers management coaching

LEADx, which is announcing its learning platform at the start of the HR Technology Conference, is using IBM Watson in its product, Coach Amanda.

Coach Amanda aims to improve managerial skills with the help of a virtual trainer. The system uses the Watson Personality Insights module, as well as its natural language conversational capabilities. The Insights program diagnoses personality to help shape the chatbot response, as well the answers and learning materials it delivers to the manager, said Kevin Kruse, founder and CEO of the firm.

Kruse said it works like this: A user can type or speak to the chatbot and ask, for instance, “What is the definition of employee engagement?” The manager may follow with a question about seeking tips on employee engagement. The chatbot answers these questions with material from a resource library based on what it knows about the manager.

The underlying IBM NLP technology has to figure out what the manager is asking about. Is the question about an employee problem? Is the manager seeking advice? Or, said Kruse, is the manager seeking a resource?

But not all firms use big vendor chatbot platforms to power HR chatbots.

HR chatbots at 2018 HR Technology Conference & Expo
HR chatbots will be in the spotlight at this year’s HR Technology Conference & Expo.

In-house and open source seen as superior by some

Jane.ai is designed to make all of a company’s information available, whether it is in a PDF or spreadsheet or resides in applications such as ServiceNow, Workday, Salesforce or among team members. HR is one of the major uses of the application, and that’s why this firm will be at the 2018 HR Technology Conference. SearchHRSoftware is the media partner for the conference.

David Karandish, founder and CEO of Jane.ai, said the system was developed in-house but also used some open source tools, such as software in Stanford CoreNLP, which provides a suite of language tools. Jane.ai developed proprietary algorithms to make matches and mine documents, he said.

An employee can use the chat system, for instance, to check vacation time or ask a question about HR policies. It can put in an IT ticket or schedule a meeting with staff.

The firm is up against the large IT vendors in AI-related development, but Karandish said the big vendor HR chatbots weren’t necessarily designed to solve a business problem. That’s why Jane.ai went with the in-house approach, he said.

“A lot of companies are coming out with cool tech, but they haven’t figured out how to actually go solve real problems with it,” Karandish said.