Uranium aerosol exposure can be a health risk factor for workers in the nuclear fuel industry. Good knowledge about aerosol dissolution and absorption characteristics in the gastrointestinal tract is imperative for solid dose assessments and risk management. In this study, an in vitro dissolution model of the GI tract was used to experimentally study solubility of size-fractionated aerosols. The aerosols were collected from four major workshops in a nuclear fuel fabrication plant where uranium compounds such as uranium hexafluoride (UF 6 ), uranium dioxide (UO 2 ), ammonium uranyl carbonate, AUC [UO 2 CO 3 ·2(NH 4 ) 2 CO 3 ] and triuranium octoxide (U 3 O 8 ) are present. The alimentary tract transfer factor, f A , was estimated for the aerosols sampled in the study. The transfer factor was derived from the dissolution in the small intestine in combination with data on absorption of soluble uranium. Results from the conversion workshop indicated a f A in line with what is recommended (0.004) by the ICRP for inhalation exposure to Type M materials. Obtained transfer factors, f A , for the powder preparation and pelletizing workshops where UO 2 and U 3 O 8 are handled are lower for inhalation and much lower for ingestion than those recommended by the ICRP for Type M/S materials f A = 0.00029 and 0.00016 vs. 0.0006 and 0.002, respectively. The results for ingestion and inhalation f A indicate that ICRP's conservative recommendation of f A for inhalation exposure is applicable to both ingestion and inhalation of insoluble material in this study. The dissolution- and subsequent absorption-dependence on particle size showed correlation only for one of the workshops (pelletizing). The absence of correlation at the other workshops may be an effect of multiple chemical compounds with different size distribution and/or the reported presence of agglomerated particles at higher cut points having more impact on the dissolution than particle size. The impact on dose coefficients [committed effective dose (CED) per Bq] of using experimental f A vs. using default f A recommended by the ICRP for the uranium compounds of interest for inhalation exposure was not significant for any of the workshops. However, a significant impact on CED for ingestion exposure was observed for all workshops when comparing with CED estimated for insoluble material using ICRP default f A . This indicates that the use of experimentally derived site-specific f A can improve dose assessments. It is essential to acquire site-specific estimates of the dissolution and absorption of uranium aerosols as this provides more realistic and accurate dose- and risk-estimates of worker exposure. In this study, the results indicate that ICRP's recommendations for ingestion of insoluble material might overestimate absorption and that the lower f A found for inhalation could be more realistic for both inhalation and ingestion of insoluble material.