Inefficiency is expensive. Just ask the city of New York.

The city’s subway system isn’t known for a glowing on-time reputation, and after a study conducted by New York City’s Independent Budget Office (IBO), we know exactly how much money those transit delays cost every year.

In 2017 alone, transit issues have caused city workers to miss more than 17,000 hours of work (and they’re on their way to missing a whopping 26,000+ before year’s end). Multiply that yearly total by the median hourly rate for a New York City-based employee ($32.40), and what do you get?

$842,400.

Nearly a million dollars of productivity wasted, because the railways are occasionally 10 minutes behind schedule.

But hey, it’s the New York subway – a complicated subterranean train system based in one of the most congested cities on the planet. How in the world would someone go about fine-tuning something so complex?

Public Transit Improvements

The answer is simple – Six Sigma. New York City transit is a network of rote systems and processes, and that’s exactly what Six Sigma’s data-driven methodology is designed to improve. It may take a while to completely overhaul a system as large as the New York railway, but consider this: if process improvement strategies can make these trains even 1% more efficient, it saves $8,400 worth of productivity.

$8,400! For a 1% improvement!

Earlier this year, an urban bus system used Six Sigma DMAIC principles to improve transportation outcomes by 20%, and the same methodology can be applied to the New York subway system.

How did they do it?

Step One: Identify Areas of Improvement – In the bus study, they systemized the actual selection of drivers. Which driver was best for which type of bus? Which driver was most suited for a specific route? Every decision they made was based in logic and reason.

Step Two: Identify Areas of Potential Risk – Every change creates risk. It’s up to you to determine if those risks are worthwhile, or if they’re severe enough to make you rethink your solutions. For example, in the bus study, matching the best drivers to the right routes might’ve uncovered a severe dearth of talent in urban bus drivers. It might frustrate or alienate the drivers, inspiring them to find jobs elsewhere. But they accepted these risks and followed through.

Step Three: Test – Once you’ve identified specific areas of improvement and risk, it’s time to get gather some hard data. Tweak and adjust those two areas, run a few tests (the study specifically cites tests like ANOVA, logistic regression, k-means cluster analysis, and others.), and record the results.

Step Four: Adjust Your Approach Based on Your Results – There’s not a one-size-fits-all solution to something as big as New York City public transportation, so it’s important to adjust testing based on the results you get.

And that’s it. Four steps.

Remember, the goal isn’t perfection. If the process can improve at all, it represents a significant gain for the city, so as long as the rate of improvement keeps increasing, you should continue testing and evaluating the results.