Industry and Academia coming together to improve maintenance outcomes with Machine Learning

DINGO is the world leader in providing Predictive Maintenance solutions to asset-intensive industries. By partnering with QUT, DINGO achieved business outcomes within 2–3 months using machine learning development to improve the scale and impact of its predictive maintenance capability. The LABS interviews Gary Fouché, DINGO Vice President of Research and Development.

The Labs
The Labs
May 9, 2018 · 5 min read
Gary Fouché, DINGO Vice President of Research and Development.

At what stage does a company decide it needs help?

DINGO delivers predictive maintenance solutions to asset-intensive companies, and this is a high-growth sector. Due to the significant global demand for DINGO’s products and services, scalability is paramount to our success. While our software was highly scalable, we knew that we needed even smarter, better technology to meet the strong demand for our products. Academia fosters the sort of thinking that brings about revolutionary technologies, so it was only natural for our business to explore a partnership with a leading university to help take our software to the next level.

Why is it sometimes beneficial to look outside for support (vs. build the capability internally)?

In cutting-edge areas of research and development, there aren’t a lot of skilled resources in the market because the technologies are so new. However, the universities often have departments staffed with professors and PhD candidates who are making significant breakthroughs in these emerging fields. By partnering with universities, companies gain access to new thinking and technology that hasn’t hit the market yet.

Also, engaging with a university allows a company to solve a tough problem by tapping into the incredible minds and diverse resources that are unique to an academic institution. This type of partnership not only provides access to a broad array of skills and resources, it helps keep costs low because you don’t have to hire full-time resources.

What problem was DINGO trying to solve?

DINGO was developing machine learning algorithms to improve the capability of its TRAKKA® predictive maintenance software and more accurately predict impending equipment failure. By using Condition Monitoring data, we analysed the patterns and correlations in the data to determine wear and degradation. These past identified patterns were then used to predict how long components of the equipment would last before requiring replacement or servicing. We knew that the optimal solution was Artificial Intelligence (AI), but DINGO had limited expertise in AI and machine learning technologies.

DINGO and QUT researchers partnered to develop machine learning algorithms to improve the accuracy of predicted equipment failure.

How did DINGO even know where to look for support?

We have been involved with the universities in our region for some time, as numerous DINGO employees are active alumni of the leading universities in Brisbane. Through these alumni connections we knew that some of the universities had deep expertise in AI, and that a partnership would help us develop a better solution faster. We also used the Advance Queensland website and other Federal government grant programs to understand partnership options and to establish connections.

Why did you decide to work with QUT?

QUT has a collaborative, outward focus, and their Knowledge to Innovation partners are motivated to establish longstanding relationships. QUT is also serious about partnerships and has the resources, research, and academic units necessary to support corporate initiatives. The university also cultivates opportunities by giving their faculty time to explore, develop and nurture partnerships.

What were the benefits of working with QUT?

QUT has some of the best researchers in the areas of machine learning and AI. QUT also ensures there is access to multiple disciplines, which is important for the types of problems we are trying to solve as they contain a combination of math, statistics, computer science and engineering. As no single faculty covers all these fields, expertise is needed across different departments and teams; QUT facilitates this access extremely well.

Did anything surprise you along the way?

The extent to which we could access the top minds, such as heads of department and professors in each field — that level of interaction was pleasantly surprising. Also, the level of engagement was fantastic. The team at QUT truly cared about the outcome of the project.

What was the outcome of partnering with QUT?

Through this partnership DINGO was able to build a collection of machine learning models to predict failure of equipment with a high degree of accuracy. These models will continue to evolve and improve over time, and we are very pleased with the way QUT experts helped us design and develop these algorithms. The outputs of the project are already being used by some of our customers to help predict the Remaining Useful Life of major components.

By partnering with QUT, our staff have also established strong professional relationships with key academic and research staff. These relationships are enduring and extend well beyond this project.

In the context of the DINGO experience, can you provide any tips or key learning that like-companies would find useful?

We were able build the relationship and progress the project at a pace that we could handle easily at the time. In hindsight, we could have engaged faster and more frequently with the QUT team.

We recommend diving into the project headfirst to get the full benefit of working with some of the best people in their field and to maximise the opportunity. We found that having our team members spend some time working out of the university offices and close to the QUT team was very helpful in developing the relationship.

Anything else you’d like to add that a reader might find interesting?

Through the Advance Queensland Knowledge Transfer Partnership Grant, we were able to receive a grant from the Queensland Government that covered two-thirds of the costs of our PhD candidate for the year. This funding was enormously beneficial to our small business and allowed us to complete a valuable project. The QUT Knowledge to Innovation partner guided us through the process of applying for the grant and helped us identify the best QUT resources to successfully tackle our business challenge.

In sum, working with QUT was a true partnership with a free exchange of ideas and concepts. QUT fosters a mindset of knowledge sharing and joint problem solving, and this project demonstrates the true power of what can be achieved through the collaboration of industry and academia.

DINGO is the global leader in predictive maintenance, providing actionable intelligence to asset-intensive industries.

Find out more about research and partnerships at QUT.


By Bree Buenen — Digital Communication Coordinator, QUT Science and Engineering Faculty.

The LABS

Learning and Big Solutions (LABS) from QUT Science and Engineering Faculty

The Labs

Written by

The Labs

The LABS

The LABS

Learning and Big Solutions (LABS) from QUT Science and Engineering Faculty

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade