An alarm sounds on the factory floor: a critical piece of equipment has malfunctioned. An engineer approaches the machine, scans its QR code, and immediately accesses visual step-by-step instructions for fixing the issue created by the people who work with the same machines every day.
This is SwipeGuide, a B2B cloud-based SaaS platform that captures and scales operational knowledge, helping teams in industrial environments to create, improve, and share instructions and standard operating procedures using mobile and wearable devices. The platform is designed to help reduce errors and downtime, improve the quality of products, and help onboard new employees – all powered by the expertise of frontline workers.
SwipeGuide Chief Technology Officer Sue Li has worked in the tech industry for over a decade, with a degree in educational technology and instructional design from Harvard University’s Technology, Innovation, and Education programme for her master’s degree. Li gained experience in visual art and UX design as well as product management and software development before joining SwipeGuide in 2019. Less than a year later, she was promoted from full-stack developer to CTO.
“I learn best through hands-on learning and just by tackling problems,” Li says. “My first year as a CTO involved learning a lot about the security and scalability of the infrastructure, data privacy, and compliance. Part of the learning process was figuring out how to ask the right questions and work with experts to solve very detailed and strategic problems.”
Harnessing the power of data for frontline workers
One of the challenges that Li contends with in her role is the ever-increasing volume of data that the platform produces, including content, feedback, usage, and behavioural data. Li is developing SwipeGuide’s new strategy to figure out how to manage it and how to put the data to work.
“The strategy that we want to go forward with is self-service analytics: how can we empower users on the factory floor so that they don’t need to rely on a data scientist or analyst to get insights? We want to have all of our data in a data warehouse as a single source of truth so that we can analyse and provide those insights to the operators. I think that’s going to be an important step towards having more robust machine learning models as well. It’s going to be very, very powerful.”
SwipeGuide is expanding its services beyond its European customers to the US and China, which presents the challenge of ensuring data privacy and compliance for an increasingly global audience.
Currently clients can access analytics dashboards on the platform that can show entities the adoption of their content, like how many instructions have been created over time by specific teams in different workspaces, and how often they’re viewed.
“We want to be able to provide better insights with those dashboards, and another part of that is embedding that data right into the platform itself: being able to see which guides have been the most popular and are the highest quality. Later on, we will be able to analyse which characteristics the highest-rated guides have, maybe something about the structure of how they are written or the structure of the images or videos. This is where machine learning will come in, to help us make recommendations to improve the quality of instructions over time.”
The human factor in Industry 4.0
Life on the factory floor is changing rapidly with the onset of Industry 4.0, the fourth wave of the industrial revolution powered by data and bolstered by automation. Tools such as SwipeGuide aim to optimise operations by minimising downtime, but the insights needed to create smarter factories must come from human expertise first.
“Our work is all about empowering the frontline worker — the biggest waste is untapped human potential. I think that the problem that we’re trying to solve is all of the silent knowledge that people have: crowd-sourcing that knowledge from all the different operators and frontline workers, and then externalising and capturing that in a standard format that is easy to share,” Li says.
Contrary to popular imagery associated with Industry 4.0 – workers replaced by endless rows of indefatigable robots – Li believes that humans will have an important role to play. “There are very few smart factories out there where there is no human intervention,” Li says, citing an example of a car manufacturer in Japan that has an automated factory for building auto parts after figuring out the step-by-step instructions necessary for robots to execute those tasks.
“With the data processing power that we have now with edge computing and cloud computing, there will be a huge shift over the next 10 to 20 years in what can be automated. But in order to reach that level of automation, we need to be able to build algorithms: some of the questions we’re asking are can we emulate the procedures, can we create an algorithm with the instructions, and how can we hook performance data into operational data?”
Branching out into wearables and augmented reality
Li and the SwipeGuide team are actively exploring other types of emerging technologies that work in harmony with the platform. The company is experimenting with wearable devices that will free up workers’ hands while they repair and service machinery.
SwipeGuide has created an Android app that can be installed on a durable industrial wearable like Realwear, which creates helmets and smart glasses built for factory settings. “We are also looking into compatibility with Google Glass,” Li says. “Wearables allow the operators to be completely hands-free when they’re repairing a machine or doing whatever they need to, which allows us to do more with voice commands.”
For more complex instructions, augmented reality can help workers understand specific gestures and motions that would be hard to describe with photos or text.
“We did a pilot with XM Reality, a remote support calling platform. If a worker gets stuck on a particular instruction, the remote experts can show users what needs to happen with augmented reality. Imagine, a worker has shared a real-time video of a part of a machine that is broken or loose, and they can see a hand on their screen making a motion or drawing a shape. Making the experience interactive can really help in situations where a course of action is very complicated and difficult to describe.”
Fostering innovation through an agile, user-centric approach
The challenges that SwipeGuide is currently facing, like integrating new technologies and developing SwipeGuide’s data strategy and machine learning models, require fostering a culture of innovation within the team. For Li, the best inspiration for new ideas comes from the people who use the platform and the solutions come from her team: “I think our customers really know the most, so it’s important to get insights from the market, the users, and the customers. We do our own usability testing — for example, we created a small competition where we created step-by-step guides for creating origami. It helped us experience the challenges that our customers face while uploading images or making changes to instructions, for example.”
Li believes that an important part of innovation is having an agile mindset, especially when it comes to software development, to measure the effectiveness of a new solution or idea.
“We track events and collect data so that we can measure solutions in relation to specific goals, like increasing visibility or usage,” she says
“If it works we keep it, and if not we create another iteration of that solution and then try again. We work in an industry with huge enterprises that use the waterfall methodology, but I think that takes away from the innovation element of being able to experiment with smaller improvements, collect and learn from the feedback, and develop new and novel ways to solve problems.”