.png){kind=icon}
.png){kind=icon}
Return
Reduce-Then-Predict or Simultaneous Reduce-and-Predict? Data-Driven Sparse Modeling for Improving R&D Efficiency
.png){kind=icon}
PRE
AI
DOI：10.1109/TEM.2025.3577580.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Efficient research and development (R&D) workflows are critical in industries where early-stage results influence downstream outcomes. This study develops a predictive model to enhance R&D efficiency for a leading integrated device manufacturer specializing in printed circuit board design. To address challenges of limited data, noise and collinearity, we apply sparse principal component analysis (SPCA) to simplify simulation data, followed by least absolute shrinkage and selection operator (LASSO) regression to predict later-stage physical testing performance. Our SPCA-LASSO model reduces prediction errors by 22%–41% compared to direct LASSO regression while offering interpretable insights for engineers. In contrast, sparse principal component regression, which integrates dimension reduction and prediction, yields higher errors and unstable factor loadings. This empirical comparison between reduce-then-predict and simultaneous reduce-and-predict approaches contributes to sparse modeling and engineering analytics, offering actionable insights for improving sequential R&D processes across high-tech industries, software engineering, construction, and other sectors where early performance predictions are critical.
Keywords:
Eye diagram
LASSO
machine learning
printed circuit boards
research and development
sparse principal component analysis (SPCA)
unsupervised learning
Journal
IF:
5.2
Papers:
376
Citations:
1.2W