Field experiment was conducted at Arba Minch during 2017 and 2018 cropping seasons under irrigation condition to determine the effect of weeding frequency and interval on weeds and bulb yield of onion. The experiment was laid out in factorial arrangement in randomized complete block design with three replications and comprised of 18 treatments: four weeding frequencies (2, 3, 4 and 5 times) and four weeding intervals (at 10, 13, 17 and 25 days) compared with weedy and weed-free checks. Results showed that all the parameters considered in this study were significant (p<0.05) among the treatments. The lowest weed density (27.33 m-2 in 2017 and 24.26 m-2 in 2018) and dry weed biomass weight (0.83 t ha-1 in 2017 and 0.82 t ha-1 in 2018) were recorded in the weed free check without statistical difference with the values received from weeding interval at 10, 13 and 17 for four and five times weeding frequency in both cropping seasons. The highest weed control efficiency (83.70% in 2017 and 82.82% in 2018) was computed in the weed free plots. In the 2017 cropping season, the highest marketable bulb yield (32.8 t ha-1) was recorded in the weed free check, which was statistically similar with the result obtained in four and five times weeding frequency at 10, 13 and 17 days weeding interval. Similar trends were observed during 2018 cropping season. Thus, three to five times frequency of weeding at 10 days weeding interval could be recommended in the study area for efficient weed management and optimization of the bulb yield and profits in onion production.