In this paper, fuzzy rule-based systems (FRBSs) are introduced for productivity estimations in aircraft final assembly lines. While linear regressions have the direct explanatory ability, they lack the capability to capture nonlinearity. With a more complex structure, other ensemble-based methods such as random forests and gradient-enhanced regression trees can provide better accuracy but sacrifice model interpretability. By adopting a multi-objective optimization, the structure of FRBSs can be simplified, leading to enhanced transparency. In addition, considering the highly customized nature of aircraft, this paper explores a potential solution for predicting productivity in cases where it is not feasible to construct independent data-driven models, specifically for modified types. Preliminary findings suggest that FRBSs show promising potential as a viable alternative compared to other available options. The adoption of multi-objective optimization, the exploration of predicting productivity for modified types, and the interpretation of fuzzy rules contribute to the advancements in both accuracy and interpretability in the field.