Tyre Tread Wear Prediction
Case Study:
Client Challenge
A specialist in the manufacturing and distribution of retread tyres. Their core business involves providing fleets and their owners with cost-effective, reliable retreading solutions that extend the life of tyres, thereby reducing overall tyre costs, as well as environmental impact.
Safe and efficient operation of heavy-duty trucks relies on proper tyre tread depth.
The client wanted to explore the feasibility of predicting tyre tread wear using their data to collect quantitative insights into the factors influencing tread wear, such as tyre design and brand, specific fleets, and tyre positions.
A key challenge was the variability in data quality, due to human measurements and the involvement of various third-party service providers.
The client ultimately wanted to leverage their data to benefit their existing, and any new customers by predicting tyre tread wear allowing them to proactively and efficiently schedule the retreading of tyres based on predicted wear rates.
The Solution
BSC built a machine-learning model to predict tyre tread wear over time, given data such as the following:
- The remaining tread depth.
- The brand and design of tyre.
- The tyre position.
- The fleet.
The model successfully predicted tread wear (with 80% of predictions to < 2.5mm error).
An interactive dashboard brought tread wear trends to life, validating the client’s industry knowledge with hard data. It even unearthed new insights that are now fueling their customer conversations.
Solution Value
Cracking the code on tyre wear! The BSC model predicted wear with pinpoint accuracy, validating the client’s experience.
This paves the way for a new data product that optimises tyre lifespans, slashing maintenance costs and boosting uptime for their customers.
Internally, it empowers them to streamline operations and maximize fleet efficiency.
- Predicted tread wea to within 2.5mm.
- Enabled the client with data-driven decision-making.