In the design and manufacture of turboshaft engines, numerous non‑probabilistic uncertainties arise under data‑scarce conditions, rendering classical probabilistic methods inadequate. This paper introduces Uncertainty Theory to establish a non‑probabilistic analysis framework for key engine parameters. Focusing on high‑pressure turbine tip clearance as a critical parameter, an evolutionary model of its uncertainty throughout manufacturing and service life is systematically developed, and the influence of clearance variations on performance is quantified in terms of belief degree. Results demonstrate that Uncertainty Theory effectively addresses tip‑clearance variability: for example, the upper bound of efficiency loss at a 95 % belief degree is predicted more conservatively and comprehensively than by traditional probabilistic approaches. Furthermore, by optimizing the nominal tip clearance from 0.50 mm to 0.52 mm, the pass rate of performance requirements under high belief degree increases markedly from 78 % to 96 %, significantly enhancing design robustness. The proposed methodology provides a rigorous theoretical tool for non‑probabilistic uncertainty analysis and reliability enhancement of turboshaft engines.