Robotic dog detects leaks for John Dewar & Sons

Robotic dog detects leaks for John Dewar & Sons

Bacardi introduces robotic dog to detect whisky leaks. The initiative aims to enhance efficiency in identifying ethanol leaks at John Dewar & Sons. Traditionally reliant on manual inspection, the new technology offers a consistent, data-driven approach. This trial could revolutionise leak detection across various industrial sectors.


Bacardi has introduced a robotic dog at its John Dewar & Sons maturation site, aiming to merge traditional craftsmanship with modern robotics. The robot, named “Royal Bark-la” in homage to the Royal Brackla single malt, is central to a pioneering trial within the Scotch whisky industry. This initiative seeks to assess whether autonomous technology can detect minor yet costly leaks in ageing warehouses more efficiently than conventional manual inspections.

In the whisky industry, the “Angel’s Share” refers to the natural evaporation of alcohol as whisky matures in oak casks. While some loss is anticipated, unforeseen leaks can lead to substantial waste. Industry-wide evaporation results in the loss of approximately 440,000 casks annually. Monitoring these barrels traditionally involves laborious, visual inspections prone to human error. By deploying a Boston Dynamics “Spot” robot, Bacardi intends to establish a reliable, data-driven alternative.

The trial is a collaboration between Bacardi, the National Manufacturing Institute Scotland (NMIS), and the Scotch Whisky Research Institute (SWRI). Engineers at the NMIS Digital Process Manufacturing Centre in Irvine have developed a custom robotic sensing kit for this project. The system includes a 3D-printed arm with sensitive ethanol vapour sensors. As Royal Bark-la navigates the Glasgow warehouse, the sensors detect vapour levels in real-time, providing consistent readings. Unlike human inspectors, the robot remains focused and undistracted by the vast environment. Angus Holmes, whisky category director at Bacardi, stated, “Craftsmanship and heritage remain at the heart of our production, but there is also great potential for innovation and technology to support the industry to become more efficient and data-driven.”

While the trial is in the “proof of concept” stage, initial results are promising. Bacardi conducted baseline testing to establish “normal” vapour levels, enabling the robot to identify anomalies indicative of leaking casks. The National Manufacturing Institute Scotland believes this technology’s potential extends beyond whisky. Autonomous detection of gas or liquid leaks could be transformative for the chemical, energy, and broader manufacturing sectors, where safety and waste reduction are crucial.

The project’s next phase may involve integrating the sensors directly into the robot’s body, enhancing durability and functionality in confined warehouse spaces. As the industry looks forward, Royal Bark-la symbolises a shift towards “Whisky 4.0.” For the team at John Dewar & Sons, the robot has quickly become a mascot for a new era — one where advanced “canines” ensure whisky loss is limited to what is destined for the bottle.


Stories for you


  • Corplex aims to expand retail offering with durable and robust Correx®-made systems

    Corplex aims to expand retail offering with durable and robust Correx®-made systems

    Corplex is targeting further retail warehouse growth with reusable bins. Made from Correx®, the modular AkyPak® Store system is being used by high street brands to improve picking stability, reduce product damage, and replace disposable cardboard with a longer-life alternative.


  • Industry urges government to accelerate skills reforms

    Industry urges government to accelerate skills reforms

    Manufacturers demand Treasury release £1bn for skills training. Almost all manufacturers identify skills supply as crucial for growth, with half viewing it as their primary challenge. The education system’s inadequacy is a concern, as digital transformation and AI demand a skilled workforce. Industry seeks urgent government action.