Glasgow, Heathrow and the resilience of our infrastructure

The huge fire that closed Scotland's busiest railway station for ten days raises a question bigger than its cause. It is the same question raised by a substation fire at Heathrow and a car park fire at Luton Airport: how resilient is our critical infrastructure to the fires that inevitably occur around it, writes Tom Roche, secretary of the British Sprinkler Alliance?

The fire on Sunday 8th March broke out in a retail unit adjacent to Glasgow Central Station. Within hours, the B-listed Victorian building had partially collapsed, businesses had been destroyed and Scotland's busiest station had closed. Full services did not resume until 18th March, ten days later.

Public attention quickly focused on the nature of the premises involved which is understandable. But it is, in an important sense, the wrong conversation. Could a general fire that started elsewhere in that building or any of the other buildings adjoining the station lead to a similar outcome? This leads you to something more fundamental: why did a fire in a single commercial unit adjacent to a major national transport hub render result in it being inoperable for over a week?

A pattern worth examining

The fire in Glasgow did not occur in isolation. This month marks the first anniversary of the North Hyde substation fire in Hayes, which cut power to Heathrow Airport on 20^th March 2025, cancelling over 1,000 flights and disrupting some 200,000 passengers. The subsequent investigation found that key equipment was overdue replacement and that the substation's fire suppression system had been inoperable since at least 2022.¹ In October 2023, a fire at Luton Airport's multi-storey car park destroyed over 1,300 vehicles in a structure that was fully compliant with current building regulations. It closed the fourth busiest airport to London and hampered passenger traffic for subsequent months.

Three fires. Three pieces of critical infrastructure. Three very different causes. In each case, the disruption was disproportionate, and in each case, questions about risk assessment, fire suppression, design standards, and resilience were raised after the fact, rather than before.

Compliance is a threshold, not a guarantee. It tells us a building met minimum requirements at a point in time but says nothing about performance under the fire conditions that actually occur in practice.

Compliance is not resilience

The Bedfordshire Fire and Rescue Service report on the Luton fire made this point directly. Current guidance still draws on fire test data from the 1960s and 70s, which does not reflect modern fuel loads, materials, or building uses. The same logic applies more broadly. The built environment of 2026 is not the one for which much of our regulatory framework was originally written for. As the Glasgow fire highlights, we use and reuse buildings multiple times in their lifetime. What once may have been true about lower fuel loads and limited fires cannot be assumed throughout time. The exposure they create to infrastructure therefore changes over time.

Effective fire suppression changes outcomes as part of the overall fire strategy. The Business Sprinkler Alliance has documented numerous cases, including car park fires in Stowmarket and at a Derby shopping centre, where automatic suppression systems contained fires that would otherwise have escalated. At Heathrow, a functioning suppression system may not have prevented the transformer fault, but it could have materially altered what followed.

When you consider the fires in Glasgow, Heathrow and Luton, each one has been treated as a distinct event with a distinct cause. Taken together, they point to something more systematic: a significant gap between the resilience our infrastructure requires and the standards our built environment currently delivers. That gap is worth closing before the next anniversary comes around.

¹ https://www.neso.energy/document/363891/download 7.17