Skip to main content

TRIGO uses artificial intelligence to improve aerospace supply chain performance.

It’s called the OTD (On Time Delivery) Predictor and it works!

The use of artificial intelligence is a hot topic. Not a single day goes by without articles and press releases explaining that it will trigger a revolution in the industry and change the day-to-day practices of manufacturing. In the Aerospace industry, some key players have made significant investments in big data analysis software. Can we really state that it has improved the performance of the industry or the supply chain? Should we consider the problem from a slightly different angle?

 

TRIGO Aerospace, Defense & Rail’s mission is to improve the performance of Aerospace and Defense supply chains, both in terms of On Quality Delivery (OQD) and On Time Delivery (OTD).

Our mission is to deliver the parts on time and at the proper quality level so that our customers can seamlessly continue to manufacture their multimillion-dollar products without a glitch. In doing so, on top of dispatching technicians and engineers to supplier sites as boots on the ground, we utilize proprietary software to collect and analyze OTD/OQD data for various manufacturing sites we support in our portfolio.

 

In developing the technology, we asked ourselves a few key questions: "Can we use AI to really improve the day-to-day life of the people in charge of supply chain performance?

Our goal is not to provide gigabytes of data analysis or multiple charts. We want to offer something simple and easy to interpret that could be used as a decision-making tool. Then we asked ourselves: “If we manage a portfolio of 50 suppliers delivering to a factory, could we predict the ones that will be late next month? If we did, could we then focus our attention, the management’s attention, and the suppliers’ attention on those key areas so that the problem is taken care of?”

 

In order to answer these questions, we developed an AI-based tool called the OTD predictor. Given a portfolio of suppliers, reviewing a neural network of the last 12-month’s history of deliveries, the tool will generate a predictability report with a list of the top 10 suppliers who have the highest probability to fail on their deliveries in the next month.

It’s as easy as that. The output is just a simple list of a few suppliers that potentially have a high probability of failure-not 200 pages of data.

 

The OTD tool’s predictability is more than 90% accurate. In fact, the more it learns, the better it is. Our first version of the tool had an 80% accuracy rate! A 90% accuracy rate means that when the tool predicts a supplier will fail, they fail in 9 out of 10 cases, so the Supplier would not be meeting its deadline requirements.

At a time when skilled people are a scarce resource, AI can help optimize the human workforce. We don’t intend to replace quality engineers; we aim to assist them in being more efficient by designating the top priority problems they should focus on and remove unnecessary work.

 

OTD predictor is a base tool that is now included in our service package along with databases, KPIs, and client portal access.