Applying spatiotemporal ICA with DEA approach in evaluating the training institution efficiency of the Semiconductor Institute program in Taiwan

In this paper, a two-stage approach of integrating spatiotemporal independent component analysis (stICA) and data envelopment analysis (DEA) is developed for efficiency measurement. stICA is used to search for latent source signals where no relevant signal mixture mechanisms are available; and DEA is used to measure the relative efficiencies of decision making units (DMUs). We suggest using stICA first to extract the input variables for generating independent components (IC), then selecting the ICs representing the independent sources of input variables, and finally inputting the selected ICs as new variables in the DEA model. The training institution dataset provided by the Semiconductor Institute in Taiwan is used for analysis. The result shows that the proposed method can not only separate performance differences between the training institutions but also improve the discriminatory capability of the DEA's efficiency measurement. The study results can serve as a reference for training institutions wishing to enhance their training efficiency.