Log Mean Divisia Index Decomposition Analysis of the Demand for Building Materials: Application to Concrete, Dwellings, and the U.K.

Dwellings are material
intensive products. To date, material use
in dwellings has been investigated mainly using economic (exogenous)
or dwelling (endogenous) drivers, with few studies comprehensively
combining both. For the first time, we identify a comprehensive set
of such drivers of demand for building materials and analyze them
using the logarithmic mean divisia index (LMDI) method. We combine
the LMDI method, the concept of dynamic material flow analysis, and
physical and monetary flows to decompose the demand for building materials
into the following six effects: material intensity, floor area shape,
dwelling type, dwelling intensity, economic output, and population.
We analyze these six effects on demand for concrete in new dwellings
in the U.K. from 1951 to 2014, classified into six dwelling types
and four subregions. Of these six effects, the material intensity
effect is the most important, overall contributing to increasing concrete
demand by +79 Mt from 1950 to 2014, while the dwelling intensity effect
plays an opposite role, overall reducing concrete demand from 1950
to 2014 by −56 Mt. The economic output effect is also significant
(+38 Mt from 1950 to 2014). A comparative analysis of the six effects
in the four U.K. nations reveals that most of the effects arise from
England, while the other nations have minor effects due to their smaller
populations. Our results show that changes to the demand for concrete
in the U.K. fluctuate and have mainly remained between ±30 Mt
year–2 from 1950 to 2014, and thus the inflows of
concrete into the in-use stock of dwellings have experienced neither
entirely increasing or decreasing trends during this period. This
study contributes to understanding changes in resource demand due
to social, economic, and technological factors and thus improves the
capability to reliably and quantitatively model the use of materials
in the built environment.

住宅属于材料密集型产品。迄今为止，针对住宅材料使用的相关研究多基于经济（外生）或住宅（内生）驱动因素展开，鲜有研究将两类因素进行全面整合。本研究首次完整识别出建筑材料需求的驱动因素集合，并采用对数平均迪氏分解法（logarithmic mean divisia index, LMDI）开展分析。本研究将LMDI方法、动态物质流分析（dynamic material flow analysis）概念与实物、货币流相结合，将建筑材料需求分解为六大效应：材料密集度、建筑面积形态、住宅类型、住宅密度、经济产出与人口规模。我们针对1951年至2014年英国（U.K.）新建住宅的混凝土需求展开六大效应分析，该分析按六种住宅类型与四个分区进行划分。在上述六大效应中，材料密集度效应最为关键：1950年至2014年间，该效应累计推动混凝土需求增长79 Mt；而住宅密度效应则发挥了反向作用，同期累计减少混凝土需求56 Mt。经济产出效应同样影响显著，同期累计推动混凝土需求增长38 Mt。针对英国（U.K.）四个分区的六大效应开展对比分析后发现，绝大多数效应贡献均源自英格兰地区，其余分区因人口规模较小，效应贡献相对有限。研究结果表明，1950年至2014年间英国混凝土需求的变化存在波动，且整体维持在±30 Mt·年⁻¹区间内，因此同期进入住宅在用存量的混凝土流入量既未呈现持续增长趋势，也未出现持续下降趋势。本研究有助于理解社会、经济与技术因素驱动下的资源需求变化，进而提升对建成环境中材料使用进行可靠定量化建模的能力。