Colin McMahon is a senior market research analyst working with PTC’s Corporate Marketing team, helping to provide actionable insights, challenging perspectives, and thought leadership on trends, technologies, and markets. Colin has been working professionally as a research analyst for many years, and he enjoys examining and evaluating just how large the overall impact of digital transformation technologies will be. He has a passion for augmented reality and virtual reality initiatives and believes that understanding the connected ecosystem of people and technology is key to a company fully realizing its potential in the 21st century.
The terms artificial intelligence (AI) and digital transformation (DX) are linked. Even when organizations or thought leaders only name one, they will likely still be referring to both working in tandem. AI, in short, already is and will continue to power the next phase of DX initiatives and software, creating opportunities and improvements not possible previously.
Since AI has myriad definitions, depending on setting and usage, we must first elaborate on its meaning as it relates to the concept of DX.
What is artificial intelligence (AI) in digital transformation?
Most often when AI is discussed in the context of engineering and manufacturing, we are actually referring to artificial narrow intelligence. It is not a matter of machines thinking like people, but rather sophisticated algorithms designed for a pre-defined task with a well understood set of inputs. Artificial narrow intelligence designed for CAD applications, for instance, will never have a “thought” outside those specific, previously outlined parameters.
Unlike standard automation, AI-powered processes can react to new information or unexpected changes. That is its biggest benefit. Unrestricted by predetermined outputs, AI algorithms learn from success and failure. They are capable of self-correction and can analyze data to detect incoming challenges before they occur.
From another perspective, automation provides optimal value when deployed in a pre-existing, well-defined process, such as established manufacturing lines. The user always dictates automation processes within the confines of a set of given inputs. AI, by contrast, is best utilized when trying to overcome challenges that are more complex, or not defined by preset rules. If the user gives an AI a set of inputs, the AI will analyze the data and suggest the optimal course of action, or just execute it automatically (depending on the situation).
Why AI is important to digital transformation
At PTC, we believe that the greatest power of digital technology is to transform the physical world – improving productivity, innovation, and impact. To that end, AI is essential for many complex DX applications. Without it, the digitization of products and processes would produce amounts of data that no human could be expected to analyze and react to within an acceptable timeframe. As a result, if you lift up the hood of nearly any PTC product, you’ll find AI powering critical applications, such as the generative design in Creo, or predictive analytics in Thingworx.
Let’s use Vuforia for an example. Vuforia is a powerful, scalable enterprise augmented reality (AR) platform – but what makes it so? Well, unlike simpler GPS-based or QR/barcode-based AR programs, which rely on data stored outside the object for AR functionality, Vuforia products increasingly use computer vision to actually identify the hardware components the user is looking at based on shape and other visual features. To do that it needs access to a database, and it needs to be able to read the patterns in the shapes to accurately identify what the user is seeing. This creates a greater level of efficiency for the user.
For another use case, let’s look at how generative design is used in CAD. It is no secret that many of today’s engineers use 3D CAD programs (such as Creo) as essential tools for creating and developing product designs. While this process is definitely faster than creating paper-based copy, it still is not necessarily optimized. For instance, many engineers are given system design requirements before they begin – and in a non-AI-enhanced program they would have to build the design from scratch. Generative design uses AI to automate this complex process, automatically generating the optimum design with minimal manual input. It’s a powerful behind-the-scenes technology that makes engineers faster and more efficient and results in innovative designs.
It is not enough to collect data from DX initiatives. Without an AI component, the likelihood of any collected information from various stages of the product lifecycle being utilized to improve efficiency or reduce breakdowns is low. In addition, organizations not currently pursuing AI initiatives within a larger DX strategy risk falling to digital laggard status. A 2021 study from PwC found 86% of its respondents identified AI as a mainstream technology. Roughly 33% have already started implementing limited AI use cases, while a quarter of respondents had fully enabled, AI-augmented processes in widespread adoption.
What are the benefits of AI in digital transformation?
Any technology, AI included, must be considered with an organization's profitability in mind. Companies today are already looking to apply DX initiatives in very controlled settings, where the outcome can be weighed against the bottom line. It isn't enough to simply say "using this DX technology makes this process better." Organizations should be measuring to identify exactly how and why their investments are impacting their workflows. This mindset removes ambiguity and allows executives to speak and act with greater confidence regarding company direction. With this approach in mind, we have identified four key, measurable benefits of AI in DX initiatives:
1. More effective decision-making
Important decisions, contrary to many films and shows, cannot rely solely on gut instinct. Even a seasoned leader needs access to any and all relevant data in order to reach the optimal conclusion. Time is always a factor, so this decision must often be reached with speed. AI can help identify and highlight important information regarding product performance, workflow optimization, and predictive outcomes. A well defined program can, for instance, run millions of simulations to calculate roughly how a new product should perform within the first six months of its lifecycle, based on the data available. With information like this in hand, decision makers can better evaluate what work still needs to be done, identify where potential pitfalls lie, and make more accurate estimations and predictions of the time and resources they will need to move forward.
2. Increased profitability
AI is not a replacement to human judgment, but can be a highly effective tool when it comes to optimizing time to value. Manufacturers are constantly faced with the need to more consistently deliver working products that meet any and all regulations within a set (and usually shrinking) time table. AI helps optimize product timelines in many ways, whether it is assisting in catching problems before they occur, running simulations, or fact-checking against existing proprietary data. By fully automating many time-consuming tasks such as these, AI software frees up human resources to be better deployed on the more cognitive aspects of product development, all while reducing the resources needed, thus increasing the profit margins.
3. Enhanced analytics
Data sets can be huge and complex, coming from various sources and geographical locations. In the past, it took serious time and personnel to culminate, process, and analyze this data. While humans should absolutely remain a vital part of the final analysis, AI can rapidly streamline every aspect of this process - delivering data results in a fraction of the time. For large manufacturers with numerous assets spread throughout the globe, AI is arguably essential to delivering actionable insights within a timely manner.
4. Holistic view of the customer
The digital world is built on data, and what that data is and where it comes from is constantly changing. In the past, tools like cookies were used to help organizations gain insight into consumer behavior. Now, however, privacy concerns and other factors have led to a steep decline in cookie usage, so companies will need new tools to better understand how their customers are behaving. AI-enhanced software can and likely will be this next iteration, helping decision makers see their customers arguably better and more comprehensively than cookies were able to do.
How AI drives digital transformation initiatives
Given these benefits, it's not a shock to see why AI already is and will continue to drive DX initiatives across a wide range of industries. To better illustrate, let's look at two specific ways AI makes a difference: automation and data analytics.
Automation
Automation has always been as good as the program controlling it. Yes, a machine can automate certain assembly processes, but the work stops very suddenly if this machine doesn't know it's connected to a larger assembly line, or which process comes next. Software has been guiding automation for some time, and AI programs like machine learning have been optimizing automation for years - and we expect this to only continue. Automation is a crucial aspect of DX, because with automation often comes visibility. Once a company has automated a process, they understand it fully, they can see the average completion time, average down time, and other crucial details.
Now, apply a well-designed AI application, such as machine learning, to better understand this data, understand where the bottlenecks are, why/when the downtime most frequently occurs, and not just for one automated system but the entire ecosystem. This information comes from software yes, but from IoT sensors, edge computing devices, service reports, and many other aspects already frequently dependent on DX technology. By implementing the right AI tools into existing DX deployments, organizations can see even more benefits.
Data analytics
As said before, AI and data analytics greatly complement one another. Important information comes from many sources, even before the product is launched into the real world. With the help of AI, companies can start to fully utilize the power of the digital thread, a connected closed loop of data that comes from a product during every stage of its lifecycle, from design inception to end-of-life service. This plethora of information provides unprecedented levels of product understanding and PLM optimization opportunities.
But data that comes from this many sources must be aggregated as quickly as possible, and AI simply can compile information faster than any human can. It may not understand what it's looking at, but AI can surface any and all relevant information to make the right decision. For one quick example, please see this video from Open AI on how data can be quickly compiled into a readable chart:
How AI works with the cloud to empower digital transformation
Frequent readers or current clients of PTC may have noticed a shift in our messaging as we promote the increasingly cloud-based nature of our software. AI is part of the reason for this shift. AI needs processing power and most organizations do not have the space on-premise for extensive server rooms. Software as a Service (SaaS) products like Onshape and Arena make special use of AI, as the larger the database, the more capable and efficient the AI becomes.
SaaS solutions, which place the vast majority of computer processing in the cloud, give the benefits of AI without so many burdens. AI is key for companies becoming more agile and more reactive—even predictive—in their problem solving. A traditionally automated solution does not need the cloud the same way AI does, but it also is not utilizing nearly the same amount of computing power to analyze the data in way that will give a competitive edge. For more information on just what impact this will make, please see this highlighted video from LiveWorx 2021:
Real-world use cases of AI and digital transformation
Decision makers understand the value of technology, but they tend to be sold on its practicality. AI is not a "couple years" solution or even "a couple months" solution, it is actionable now and many organizations are already using it to augment their offerings and improve internal workflows. Here are three quick examples of AI working within DX initiatives to improve operational efficiency:
1. Customer service
Chatbots are nothing new and remain the most common form of AI customer service many people think of. Chatbots can be effective but they are often extremely limited, simply matching a customer query to one of many preset answers. These presets are often made from the most common Q&A, so they can solve a great deal of customer questions, but nowhere near all. Large language models—such as ChatGPT—by contrast, are more advanced.
These new models do not rely on any presets. Instead, the value of a large language model is often tied to the information it is pulling from. ChatGPT, while impressive, cannot tell truth from fiction, so its dependability is directly linked to where its model comes from. A large language model pulling from the entire internet may not be helpful, as there could be conflicting and false information present to dilute actual answers. That said, this new form of generative AI can be tooled to much more specific language models. For instance, PTC could create a ChatGPT-like program that only draws information from PTC and related websites. This means it is only scanning for information from approved sources, and its knowledge is far more likely to be accurate.
This new wave of chatbots will be able to react to the customer and answer them directly. It still isn't a perfect system, but it is a significant step up from earlier preset-based chatbot models.
2. Manufacturing
Manufacturing in its broadest sense has many segments and many areas when AI can be applied, so for the sake of brevity, we'll focus in on smart manufacturing. Smart manufacturing, as its name suggests, is the application of smart, connected technology (including AI) to facets of traditional manufacturing. For instance, failure prediction: Understanding when and where machinery will break down to better equip and prepare technicians. Without failure prediction, organizations are purely reactive to downtime, a stance which often lengthens the period of inactivity and exacerbates the costs associated with every single failure.
While humans are certainly capable of calculating failure predictions, it is an intensive process. AI, in smart manufacturing, can instantly read and analyze the data and offer human operators strong indicators of when and where machine failures will occur. This lets companies be more strategic and more proactive in maintenance operations, and greatly increases first-time-fix rates. Of course, AI goes far beyond this role in smart manufacturing and, to see more, please have a look at this PTC webinar on AI in the production space:
3. Healthcare
Preventative maintenance saves dollars in manufacturing. Advanced, AI-powered, healthcare analytics can save lives and improve preventative healthcare—solving problems before they become life-threatening. AI can be used to analyze millions of X-rays in seconds, helping to identify issues that even skilled technicians might miss. The same is true with CAT scans, ultrasounds, and essentially every other piece of data. All of it can be analyzed against a much larger set of results, helping doctors detect patterns and formulate treatment strategies with higher levels of information.
What is the future of AI in digital transformation?
The effects of AI in DX efforts are already felt today at multiple access points. Engineers designing CAD files with generative AI can see automatic updates in their design parameters, which opens up new design possibilities, including viable alternatives not considered before—but, when used, may be lighter, reduce material cost and save on part construction and deployment.
The executive trying to improve efficiency across her multi-location organization has access to analytics offered by the AI platforms, rather than just a surplus of data. With it, she can better execute an AI enterprise strategy. Improvements include greater visibility into company initiatives (either at the corporate or departmental level), which can accelerate the approval and production processes for new products and solutions, thus shortening time to market without bypassing key steps.
Going forward, PTC expects to see more companies embrace AI in their DX initiatives to maintain a competitive advantage. AI is integral to some of our most exciting products and will continue to power PTC solutions for years to come. This is not about replacing people with computer software. It is about efficiently analyzing and acting upon cloud-based data, giving people the tools they need to succeed today and tomorrow.
Build Better Predictive Models
Enable predictive maintenance with industrial AI
Click Here