Brassica juncea is a major oilseed crop of substantial economic and nutritional significance, extensively utilised in food systems, edible oil extraction and diversified agricultural production. Improving its productivity remains a primary breeding objective, particularly through the development of an ideotype that integrates high yield potential with resilience and broad adaptability. The present investigation combined agronomic evaluation with genetic analysis to promote sustainable mustard improvement. Conventional multi-trait selection procedures, such as the Smith-Hazel selection index, are often constrained by multicollinearity among traits and reliance on subjective economic weights. Therefore, the multi-trait genotype-ideotype distance index (MGIDI) was adopted as a more flexible and statistically robust alternative. The experiment was conducted at the Agricultural Research Farm, BHU, Varanasi, during the rabi seasons of 2019–20 and 2020–21 using a randomised block design with three replications, where sixteen agronomic and yield-related traits were recorded. Combined analysis of variance revealed significant genetic variability among the mutants, while genotype × environment interaction was largely non-significant except for seeds per siliqua. Principal component analysis was performed to reduce the data complexity, classifying traits into five principal factors explaining 79.6 % of total variation. Application of MGIDI resulted in a cumulative genetic gain of 85.3 %, with notable improvement in secondary branches and siliqua per plant. Four mutants, TM-130, TPM-1, TM-53 and Kranti were identified as superior genotypes for future breeding, underscoring MGIDI’s effectiveness in multi-trait mustard improvement programmes.
