The 90-Second Power Cut That Paralyzed a City for Days Is Not Bad Luck. It’s a Design Flaw.

You’re standing on a platform, phone battery at 12%, watching the departure board flip from “Delayed” to “Cancelled” for the third time this week. You don’t know why. You don’t need to know why. All you know is that the invisible machinery that supposedly runs your life has silently coughed, and now you’re going to be ninety minutes late for something that matters.

Then you read the news: the entire disruption was caused by a 90-second power cut. Ninety seconds. Less time than it takes to microwave a burrito. And it’s now rolling into day two of chaos.

Let that sink in.

A system that moves millions of people every day — with schedules timed to the minute, with safety systems that are supposed to be fail-safe, with backup plans and redundancy strategies — was knocked off its feet by a blip that lasted roughly the length of a single traffic light cycle.

And the response from the people running it? “We’re working as fast as we can.”

That is not a failure of operations. That is a confession of architecture.

We’ve built a world where a 90-second blip can wipe out two days of your life. That’s not resilience. That’s a house of cards.

Think about how a modern rail network works. It’s not a collection of independent trains driven by autonomous humans. It’s a tightly coupled system: every train’s position is tracked, every signal is coordinated, every crew roster is optimized down to the minute. The moment something deviates even slightly, the system enters a cascade of re-planning. A power cut of any length disrupts the signals, the points, the communication backbone. But in a system with slack, that would be a hiccup — trains would slow, wait, recover. In a system with zero slack, the hiccup becomes a cascade.

And so a 90-second power cut becomes a 90-minute delay for one train. That delay pushes it out of its scheduled slot. That slot was already occupied by the next train. The crew for that train is now over hours. The rolling stock that was supposed to be at the depot for maintenance is now sitting in a tunnel. The maintenance crew’s shift ends. The next day’s schedule is built on the assumption that everything recovered overnight — but it didn’t, because the trains are in the wrong places and the crews are exhausted.

This is what engineers call a "cascade failure." But it’s not a rare event. It’s the inevitable outcome of a system that has been optimized for efficiency, not resilience. Every year, we shave another minute off the timetable. Every year, we run trains closer together. Every year, we cut the "buffer time" that used to exist precisely to absorb these tiny shocks. And then we act shocked when a small shock takes down the whole house.

The problem isn’t the power cut. The problem is that we’ve normalized the fragility.

Look at the way this is being reported. The headlines say “Power cut causes train chaos.” They don’t say “Underinvestment in redundancy exposes systemic brittleness.” They don’t ask why a 90-second failure can have a 48-hour impact. They treat the power cut as the villain, when the real villain is the absence of slack.

You’ve seen this pattern before. In supply chains, where a single port closure in a distant country empties supermarket shelves. In cloud computing, where a five-minute outage at a single provider takes down half the internet. In power grids, where a single substation failure blackouts a region. We keep building systems that are so lean, so just-in-time, so optimized that they have no room to breathe. And then we’re surprised when they suffocate.

This is not about trains. It’s about the invisible logic that governs everything around you. Your commute, your food supply, your electricity, your internet — they’re all running on the same assumption: that the components will never fail. But they do fail. And when they do, the system has no idea what to do because it was never designed to fail gracefully.

What does graceful failure look like? It looks like a system that has spare capacity: extra trains, extra crews, extra time in the schedule. It looks like a network that can reroute, not just re-plan. It looks like a culture that accepts that things will go wrong and plans for it, rather than a culture that prays they won’t and panics when they do.

We need to stop treating outages as anomalies and start treating them as inevitable test cases. A mature infrastructure should absorb a 90-second fault without flinching. The fact that this one didn’t is not bad luck. It’s a metric — a direct measure of how much we’ve invested in resilience versus how much we’ve invested in squeezing every last drop of efficiency.

Resilience isn’t about never failing. It’s about failing without breaking everything else.

So the next time you’re stuck on a platform, watching the board flip to “Cancelled,” don’t just curse the power cut. Ask yourself: what else in this city is running on the same fragile assumption? The bridges? The hospitals? The water supply? The data centers that hold your bank account? The answer should unsettle you.

Because the 90-second power cut wasn’t the story. The story is that we’ve built a world that can be derailed by a blink — and we’ve decided that’s acceptable.

It isn’t.