This paper aims to identify the main driving force for changes of total primary energy consumption in Beijing during the period of 1981-2005.Sectoral energy use was investigated when regional economic structure changed significantly.The changes of total primary energy consumption in Beijing are decomposed into production effects,structural effects and intensity effects using the additive version of the logarithmic mean Divisia index (LMDI) method.Aggregate decomposition analysis showed that the major contributor of total effect was made by the production effect fol- lowed by the intensity effect,and the structural effect was rela- tively insignificant.The total and production effects were all posi- tive.In contrast,the structural effect and intensity effect were all negative.Sectoral decomposition investigation indicated that the most effective way to slow down the growth rate of total primary energy consumption (TPEC) was to reduce the production of the energy-intensive industrial sectors and improving industrial en- ergy intensity.The results show that in this period,Beijing’s economy has undergone a transformation from an industrial to a service economy.However,the structures of sectoral energy use have not been changed yet,and energy demand should be in- creasing until the energy-intensive industrial production to be reduced and energy intensity of the region reaches a peak.As sequence energy consumption data of sub-sectors are not available, only the fundamental three sectors are considered:agriculture, industry and service.However,further decomposition into secon- dary and tertiary sectors is definitely needed for detailed investi- gations. This paper aims to identify the main driving force for changes of total primary energy consumption in Beijing during the period of 1981-2005.Sectoral energy use was investigated when regional economic structure changed significantly. Changes in total primary energy consumption in Beijing are decomposed into production effects, structural effects and intensity effects using the additive version of the logarithmic mean Divisia index (LMDI) method. Aggregate decomposition analysis showed that the major contributor was made by the production effect fol- lowed by the intensity effect, and the structural effect was rela- tively insignificant. The total and production effects were all posi- tive. contrast, the structural effect and intensity effect were all negative. Sectoral decomposition investigation that that the most effective way to slow down the growth rate of total primary energy consumption (TPEC) was to reduce the production of the energy-intensive industrial sec tors and improving industrial en- ergy intensity. These results show that in this period, Beijing’s economy has undergone a transformation from an industrial to a service economy. However, the structures of sectoral energy use have not been changed yet, and energy demand should be in creasing until the energy-intensive industrial production to be reduced and energy intensity of the region reaching a peak. a sequence energy consumption data of sub-sectors are not available, only the fundamental three sectors are considered: agriculture, industry and service. However, further decomposition into secon- dary and tertiary sectors is definitely needed for detailed in-gations.