ANN-based decision model for the reuse of vacant buildings in urban areas
Because of global urbanization and sustainable development trends, reusing vacant buildings is a crucial strategy employed in urban development and management. Reusing and adjusting the future service values of unused buildings to extend building life cycles is a sustainable approach that benefits society, the economy, and the environment. However, repurposed spaces are easily re-discarded because a comprehensive system and operational plan for assessing the effects of building reuse remains unestablished. The research framework adopted in this study was based on the seven factors of the AdaptSTAR model; assessment criteria for building reuse were then created. In addition, 62 types of reused building cases in Taiwan were investigated and a decision model for reuse type prediction and business strategy was constructed on the basis of artificial neural networks. The results indicated that the proposed decision model yielded a reuse type accuracy of 89% and a business strategy accuracy of 78%. This systematic approach can be adjusted according to local conditions and applied as an effective decision support tool.