Recentl y, inter est in functional safety has surged because vehicle technology incr easingl y r elies on electr onics and automation. Fail- ure of certain system components can endanger driver safety and is costly to address. The detection of abnormal data is crucial for enhancing the r elia bility , safety , and efficiency . This study introduces a novel anomaly-detection method of designable generative ad- versarial network anomaly detection (DGANomaly). DGANomaly combines the data augmentation method of a designa b le generati v e ad versarial netw ork (DGAN) with a generati v e adv ersarial network anomal y-detection data classification tec hnique . DGANomaly not only generates virtual data that are challenging to obtain or simulate but also produces a range of statistical design v aria b les for nor- mal and abnormal data. This approach enables the specific identification of normal and abnormal design variables. To demonstrate its effecti v eness, the DGANomal y method w as applied to an electric power steering (EPS) model when multiple de gr adations of gear stiffness, gear friction, and r ac k displacement w er e consider ed. An EPS model w as constructed and validated using simulation pro- gr ams suc h as Pr escan, Amesim, and Sim ulink. Consequentl y, DGANomal y exhibited a higher classification accuracy than the other methods, allowing for more accurate detection of abnormal data. Additionally, a clearer range of statistical designs can be obtained for normal data. These results indicate that the statistical design variables that are less likely to fail can be obtained using minimal data.