Past research on air quality within the Kathmandu Valley indicates that diesel vehicles make a substantial contribution to the ambient pollution. Hence, it’s important to identify cost-effective measures for reducing their emissions. As a first step, roadside observations of diesel vehicles were recorded between February and April 2017 at six locations: two on the ring road (RR), two inside the RR, and two on major arterial highways outside the RR. Out of all diesel vehicles observed (n = 12,039), 35% were emitting a visible plume of black smoke and hereafter are referred to as “superemitters”. Of the 4,248 superemitters, 45% were buses of varying sizes, 34% were large trucks, and 19% were small pickups. Superemitters made up the largest fraction of diesel vehicle traffic on the RR (43%–46%) but were also abundant inside the RR (27%–29%), where human population and pollutant exposure is greatest. Upon developing a comprehensive understanding of the superemitting vehicle types and ownership, maintenance patterns and servicing costs were studied through a survey of vehicle owners, vehicle drivers, and local maintenance centers. The costs of general servicing ranged between USD 16 for tractors and USD 203 for construction vehicles depending on the size of the vehicle. Lastly, the effect of general servicing on emissions while idling was explored for a small sample of superemitters (n = 4). PM2.5 emissions reduced from 10.90 g L–1 to 3.76 g L–1 and BC emissions reduced from 0.847 g L–1 to 0.596 g L–1 after servicing. Taken together, results from this roadside surveillance study and exploratory emission-measurement campaign provide preliminary evidence that a policy of mandatory, routine maintenance of a targeted subset of the diesel fleet can systematically reduce emissions and improve air quality in the Kathmandu Valley and other cities around the world that are facing similar problems.