Knowledge-based Minimization of Railway Infrastructures Environmental Impact

Life Cycle Assessment (LCA) and intelligent data analysis can help in reducing carbon and water footprints of rail infrastructure construction projects. The goal is to improve the railway construction processes with regard to their environmental impact, mainly in those aspects related to climate change such as carbon and water footprints. Based on a set of 27 indicators of every task in the construction process, a comprehensive compilation of basic information was performed, where the main project units and their sub-tasks were reviewed and analyzed. Afterwards, focus was on analyzing the transformation from environmental impact to carbon and water footprints, by means of the development of a consolidated evaluation methodology. A tool is being developed based on data mining and computational intelligence approaches. It will allow knowledge-based alternative project units scheduling, conditioned to previously selected specific footprint values and environmental indicators. This decision support system (DSS), based on multi-criteria and multi-objective intelligent optimization algorithms, will help to reduce carbon and water footprints of rail infrastructure construction projects by around 10% and 5%, respectively. Tests are going to be performed on two real high speed railway construction projects. That way, a global search procedure provides an analysis of the best alternatives in the scheduling and execution of the project units and their environmental impact offering a front of solutions displaying different trade-off amongst several ‘footprints’. Results will allow the development of a series of environmental impact indicators, which will support rail infrastructure construction companies becoming more sustainable and efficient by minimizing their environmental impact.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01608171
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jun 29 2016 8:27AM