Fuel Consumption as a Spatial Proxy: Measuring Urban Infrastructure Performance in Juba, South Sudan

Imagine trying to figure out how bumpy a road is just by listening to a car engine. In the world of city planning and engineering, this is usually done with high-tech sensors, satellite maps, and expensive traffic data. But what if you live in a place where those fancy tools don't exist? This is the reality for many fast-growing cities in the developing world, where road maps are often just guesses and traffic data is missing. To solve this, scientists sometimes look for "proxies"—simple, everyday things that act like a stand-in for complex data. One such idea is looking at how much fuel a car burns. Usually, we think of fuel consumption as a measure of how efficient a car is, like how well a bicycle rider pedals. But in a city with bad roads, fuel use tells a different story: it measures how much the road itself is fighting against the car. If a car sips fuel on a smooth highway but gulps it down on a short city trip, it's not the car's fault; it's the road's. This paper asks a simple, curious question: Can we use a car's fuel tank as a secret spy to tell us how well a city's roads are actually working?

The story begins in Juba, South Sudan, a city where data is scarce and roads can be a challenge. The author, Adhar Machar Malok Machar, was on a spontaneous trip to a cultural festival when something strange happened. Driving 65 to 70 kilometers on a smooth highway to Terekeka, the car's fuel tank barely seemed to drop. It was so surprising that the author decided to skip the festival, turn the car around, and drive straight back to Juba to test a theory. The idea was to see if the amount of fuel a car uses per kilometer could act as a "spatial friction proxy." Think of "spatial friction" like the resistance you feel when trying to run through deep water versus running on a track. In a city, this friction comes from potholes, stop-and-go traffic, and rough dirt roads that force the engine to work harder.

To test this, the author took a 2010 Nissan Juke on a series of "transect" runs, which is just a fancy way of saying driving back and forth along specific paths. They chose four different routes to compare:

1. Route A: A long, smooth highway between Juba and Terekeka (140 km).
2. Route B: A short, 100% dirt road (marram) inside the city (18.2 km).
3. Route C: A mixed road inside the city with some dirt and some paved sections (11.5 km).
4. Route D: A paved road inside the city that was full of potholes (12.8 km).

The team drove each route three times back and forth to get reliable numbers, measuring the fuel drop using the car's dashboard range indicator. They calculated a "Normalized Fuel Consumption" (NFC) score to see how much fuel was burned per kilometer on each path.

The results were a big surprise. Even though the highway trip (Route A) was the longest and the fastest, it was the most fuel-efficient. The car burned a tiny amount of fuel per kilometer, with an NFC score of just 0.27. It was like gliding on ice.

In contrast, the short trips inside the city were fuel monsters. Despite being much shorter distances, the city routes burned between 6.9 and 10 times more fuel per kilometer than the highway.

• Route C (Mixed Surface): This was the worst offender, with an NFC of 2.70. Even though it wasn't the longest or the most broken road, the fact that it switched between dirt and pavement made the car work the hardest.
• Route B (100% Dirt): This had an NFC of 2.07.
• Route D (Potholed Pavement): This had an NFC of 1.85.

The study suggests that it's not just about whether a road is dirt or pavement; it's about how consistent the ride is. The mixed road (Route C) was the most inefficient because the constant switching between road types disrupted the driver's rhythm and the engine's efficiency, creating more "friction" than a single type of bad road. Additionally, roads with more intersections and stop-start traffic burned more fuel.

The author translates this into money to show the hidden cost. Using an estimated fuel price of USD 1.8 per liter, a round trip on the short, bumpy Route B (18.2 km) cost about USD 4.52. A round trip on the mixed Route C (11.5 km) cost about USD 3.72. Compare that to the 140 km highway trip, which also cost roughly USD 4.52. This means residents are paying almost the same amount of money to travel a tiny fraction of the distance because the city roads are so inefficient.

The paper concludes that fuel consumption is a powerful, low-cost tool for measuring how well a city's infrastructure is performing, especially in places where we don't have fancy traffic data. It suggests that fixing small, specific problems—like smoothing out mixed-surface roads or organizing busy intersections—could save residents a lot of money and energy. However, the author is careful to note that this is an exploratory study. They admit that using just one car (the Nissan Juke) and reading the dashboard (rather than measuring fuel with a precise cup) has limits. They also note that traffic and weather change, so these numbers are a snapshot of that specific time. But the main takeaway is clear: in a city like Juba, a car's fuel tank is a loudspeaker telling us exactly where the roads are hurting the most.