Northwest China (NWC), as a typical arid and semi-arid region, has exhibited a pronounced warming and wetting trend since the 1980s. However, changes in the precipitation structure in the NWC remain poorly understood. Based on observational data and CMIP6 multi-model simulations from 1961 to 2014, this study analyses variations in precipitation structure over the NWC and conducts the attribution analysis using the optimal fingerprinting method to quantify the relative contribution from different external factors. The projection of changes in precipitation structure is also studied. The results demonstrate a transition in precipitation structure characterised by reducing light precipitation frequency and increasing amount and frequency of heavy precipitation. Attribution analysis identifies greenhouse gases (GHG) forcing-primarily driven by anthropogenic emissions-as the dominant driver of these structural changes. In contrast, aerosols (AER) forcing and natural (NAT) forcing contributions remain statistically insignificant. The characteristics of changes in the precipitation structure in the NWC are projected to continue in the future, with a marked increase in both the frequency and amount of heavy precipitation events. Compared with the raw projection, the constrained projections exhibit a substantially stronger magnitude in precipitation structure changes. These findings underscore the urgent need to mitigate greenhouse gas emissions and to implement adaptive strategies to address heavy precipitation in the region.