Environmental and Sustainability Indicators (Dec 2023)

A review of integrated multicriteria decision support analysis in the climate resilient infrastructure development

  • Parfait Iradukunda,
  • Erastus M. Mwanaumo,
  • Joel Kabika

Journal volume & issue
Vol. 20
p. 100312

Abstract

Read online

Roads, bridges, sewer systems, and other infrastructure failures often result from climate-related incidences along with extensive socioeconomic impacts including human life losses. Infrastructures with frequent experiences are usually replaced, altered, or adopt different Low-Impact Development, Best Management Practice (LID-BMP) approaches. Adapting and recovering from the damages cost a substantial budget. This study reviewed Multicriteria Decision Analysis (MCDA) i.e., risk analysis, hydroclimatic analysis, and Life Cycle Cost-Benefits Analysis (LCC-BA) in the Climate-Resilient Infrastructure Development (CRID). It was carried out following the protocols and guidelines for systematic literature review. Throughout the review, 1D is acclaimed as the best and most suitable to identify critical flooding zones and nodes within the drainage network. The study showed that integrated GIS and 1D hydrodynamic modelling are reliable in waterlogging characterisation and locating suitable floodwater regulating areas, while 2D analysis was found ideal for appropriate damages assessment over different inundation depths, duration, return periods and different climate scenarios. Indeed, about 62.5% of the studies have analysed the LID-BMPs whereby 23.2% integrated hydrologic-hydrodynamic and LCC-BA, and identified optimum performances at different levels. The study showed that, the cost of climatic adaption in infrastructure development results in the benefits optimisation and the effects-attributed cost minimisation. Besides, several studies acclaim the rising of weather-related extremes due to a gradual climate variation. Henceforth, there is a need for adaptation, most importantly, incorporating the changes in infrastructure development, and the necessity of integrating MCDA in CRID. Further, machine learning and deep learning approaches are recommended to overcome challenges and limitations associated with the current multi-dimensional numerical models and big data era demanding huge time and computational power.

Keywords