ThingWorx Review: Ditching the Duct Tape
Many companies are going through the transition from developing mechanically oriented products to smart, connected ones. Overall, that transition includes a sea of changes, many of them challenging. This includes adding the organizational capability to develop embedded code, electrical systems, or rigid-flex electronics. One of the easier adjustments, however, is implementing an Internet of Things (IoT) platform and connecting a product to it. In some ways, this has become a plug-n-play type of change. In this post, we’ll be diving into one of the prominent IoT platforms available: ThingWorx. We’ll review the capabilities it provides and the value that organizations can obtain from it. Let’s get started.
If there’s one thing you should know upfront, it is that ThingWorx is a very broad offering. PTC, the company that provides this offering, has acquired numerous companies to add to its platform and has continued to make significant development investments over time.
Streaming and Storing Product Data in the Cloud
One of the foundational steps in making a smart, connected product is connecting to an IoT platform that sits in the cloud. This allows sensor readings and other data to be streamed to the IoT platform, where lots of interesting things can happen, which we’ll cover soon. This connection allows information, data, and even embedded software updates to be streamed back to the product as well.
The initial ThingWorx acquisition provided many capabilities. However, PTC expanded upon these capabilities when they acquired Axeda, which folded in more connections and partnerships. With this expansion, device clouds such as Azure, AWS, and GE Predix are supported. Overall, a broad base of connection capabilities allows companies to get their products talking to ThingWorx relatively easily and quickly.
Of course, once you stream your product’s data to an IoT platform, you’ll want to store it there, perhaps for a significant amount of time. ThingWorx is a cloud-based solution, but there are options. You can install ThingWorx on a single dedicated server that your company maintains in their data center, or you can set it up in a cloud service like AWS or Azure. Note that in either case, you can set up a multi-tenant environment, where many different companies can access ThingWorx on the same server. There are lots of options to tweak, depending on your goals.
These capabilities are provided by ThingWorx Foundation.
If you’re in more of an industrial environment, then you should take a hard look at ThingWorx Industrial Connectivity, which provides connections based on the Kepware technologies that PTC acquired.
Doing Something with Your Product Data
OK. So you now have product data streaming to the cloud. Now what?
You can, of course, do some analysis of your product data. However, that is an extremely simple case. More often than not, you’ll need to manipulate and transform your data to create derivative measures that actually track what you want. That is where the ThingWorx Application Enablement Platform comes into play. It includes:
- Model Composer: This tool lets you build a model that can manipulate and transform your product data. More specifically, it treats your product data as input and manipulates it. Note that this might include items like workflows that automate tasks based on logical conditions or even pull data and information from other enterprise systems like ERP, PLM, CRM and SCM.
- Anomaly Detection: Whenever something looks irregular with the data streaming from your product, you’re probably going to want to know about it sooner rather than later. The Model Composer includes analysis that can detect when the data is wildly erratic or varies far from the norm. With this detection functionality, you can build in notifications or other actions.
- Runtime Environment: Under the hood, that Model Composer is essentially developing software code behind the scenes. That code is then run in real time as data is streamed from your product. This allows those transformations, manipulations, and workflows to run just as fast as the Model Composer receives data.
All of these capabilities are also provided by ThingWorx Foundation.
App-ifying Access to Smart, Connected Products
Want remote access to your smart, connected product? Well, there’s an app for that. This might sound cliche, but it’s true.
ThingWorx supports the creation of apps that can connect to your smart products. It has an item called a Mashup Builder that is something akin to a visual programming environment. Based on the model you’ve built, you can codelessly drag and drop different components into an app. That can then be presented as a web application, a web dashboard, or an app for a mobile device.
The key here is that the Mashup Builder can be used by practically anyone. So, if someone in the service department needs to build out an app that can truly make a difference, they don’t need to worry about calling in corporate IT and waiting weeks, if not months, while it is being coded. The same is true when you want to make changes to that app.
Getting the Edge Involved
Lots of companies are talking about the edge and how important it is for IoT. But what’s involved? Why is it important?
Edge computing for IoT is basically a solution for network bandwidth. You see, if you’re streaming data from a bunch of sensors simultaneously from the same wireless antenna, you can easily surpass the antenna’s network bandwidth maximum. That’s when you start dropping readings. And in critical circumstances, that can be bad. Products might be sending you data that would ultimately raise red flags, but if you’re not receiving them, then you can’t shut down that product before it fails. Catastrophically.
Instead of sending all that data to the cloud for processing, why not stick a mini-computer on the product and process that data locally? If anomalies show up, then that message is sent to the IoT platform. However, you may need something more than just a mini-computer to stick on the product. You need your IoT model and its runtime environment to run locally, instead of in the cloud. This capability is part of ThingWorx Foundation. Note that the entire platform can be run at the edge. That means you don’t send any data anywhere. It is a use case that is applicable.
Specifically, Edge Microserver is a pre-built agent, which is the combination of a model and its runtime environment, that you can drop onto the product’s compute resource. Alternatively, you can build your own stuff using the Edge AlwaysOn SDK. Either way, ThingWorx Foundation supports an edge strategy.
Getting Deep Insights from Product Data
While it isn’t as big of a buzzword now, big data is still a problem. I’ve known many companies that have instrumented their products with a ton of sensors and started capturing monumental amounts of data. There is definitely an issue with storing all of that information in terms of network connectivity. But the bigger problem is taking the time to meaningfully analyze that data. In some cases, you literally don’t have enough time to look at that data manually. This is where machine learning comes into play.
This form of artificial intelligence uses statistical algorithms to see patterns in data. In that way, the software learns to recognize precursors in the data, allowing it to predict what will happen next. In the context of IoT and smart, connected products, that could means a huge variety of things. Most obviously, the software can predict the failure of a product in the field so that you can proactively service the product, allowing companies to deliver zero downtime. Clearly, this concept can be applied to a huge range of applications. It might mean proactively changing the parameters of a control system in order to more tightly control the ranges of product operation.
Manual and automated analytical capabilities are delivered through ThingWorx Analytics.
There’s more to the ThingWorx platform. There’s ThingWorx Navigate, allowing a user to view information from multiple enterprise systems in one place, which is more applicable to engineering than you might think. There’s ThingWorx Asset Advisor, which is a pre-packaged application to remotely monitor connected equipment. There are a number of Augmented Reality (AR) and Virtual Reality (VR) solutions, including Vuforia Engine, Vuforia Studio and Vuforia Chalk. I’ll write about a number of these solutions in other posts but exclude them here because they don’t directly relate to connecting your smart product to an IoT platform.
Commentary and Analysis
So what exactly is my take?
Ditching the Duct Tape
I think that most folks today would agree that we’re early in the IoT era. Now, the adoption of any revolutionary type of solution looks similar. You have pioneers that are willing to forge ahead and explore the possibilities of a new technology. They’ll stumble now and again as they test the new technology. But they’ll duct tape it together to make it work. At the end of the day, they’ll get some tremendous differentiated value out of that technology. At the end of the day, it’s all part of the deal. You have to ingeniously duct tape stuff together to get some awesome payback.
As I look at ThingWorx, however, I don’t see the need for almost any duct tape. Yes, of course it’s not perfect. But it is far more than serviceable. After evangelizing IoT for the past five years, PTC now has a very capable and well-integrated suite of IoT tools. What’s my point?
A lot of companies can take advantage of IoT without being a pioneer.
Seriously. If companies can figure out how to gain a competitive advantage from IoT, then they have a pretty easy path to getting the technology set up. Essentially, they get to realize the benefits of being a pioneer without paying the price of being a pioneer. That leads me to my next point.
Figure It Out. Now.
Overall, I see the majority of companies, but not all, in one of two groups.
One group is charging forward with IoT. Yet, they’re doing this from a technology perspective. Some companies have figured out what IoT means to them from a business perspective, but many have not. They’re experimenting with IoT as a means to be technically ready when they do figure out the business side of things.
The other group isn’t doing anything. Nothing. They’re frozen between opportunity and risk. The opportunity behind hooking smart, connected products to an IoT platform is obvious. Yet, these companies are scared into inaction because of the potential disruption of that change.
My message? You need to figure it out. Seriously. Now. A mass of companies is exploring what IoT means to them. Sooner or later, one of your competitors is going to figure out. Or, frankly, a startup is going to figure it out. By then, it could well be too late. Industries will be disrupted. Hope yours will be one of the industries that is disrupted last.
Why put this message in a review of ThingWorx? Because the “technology isn’t ready” excuse simply isn’t valid. The technological barrier is now really low. You have to figure out to the business side of IoT. And soon.
- ThingWorx is PTC’s IoT platform.
- ThingWorx Foundation offers a wide variety of capabilities covering:
- Connecting products to the platform,
- Storing product data in the cloud,
- Building and running a model to transform product data,
- Codelessly building apps to link to the IoT and smart, connected products,
- Running models and processing data at the edge,
- Leveraging machine learning on top of product data.
- This solution set is mature, allowing most companies to build out a strategy for IoT and smart, connected products without having to hold things together with duct tape.
- Companies need to figure out what IoT and smart, connected products mean to them from a business perspective. Soon. The technological barriers are incredibly low, making it too easy to disrupt current markets.
All right, folks. That’s my take. Let me know your feedback in the comments. Looking forward to hearing from you.