Abstract. The focus of this investigation was to understand the design space to achieve comparable in vitro performance of two multi-unit dose dry powder inhalers (DPIs)-Flixotide® Accuhaler® (reference product) and MultiHaler® (test product). Flow field, pressure drop and particle trajectories within the test and reference DPI devices were modelled via computational fluid dynamics (CFD). Micronized fluticasone propionate (FP) was characterized to determine particle size distribution (PSD), specific surface area (SSA) and surface interfacial properties using cohesive-adhesive balance (CAB). CFD simulations suggested that the pressure drop and airflow velocity in the MultiHaler® were greater than Accuhaler®. Two modified test devices (MOD MH 1 and MOD MH 2) were manufactured with the introduction of by-pass channels in the airflow path, which achieved comparable specific resistance and airflow path between the test and reference devices. Assessment of reference product formulation in modified test devices suggested that MOD MH 2 achieved comparable in vitro performance to the reference product. CAB analysis suggested that adhesion of all FP batches to lactose was different, with batch D showing greatest and batch A least adhesion to lactose. Test DPI formulations were manufactured using four different batches of FP with milled or sieved lactose, and showed that batch A FP formulated with sieved lactose in MOD MH 2 device demonstrated the highest degree of similarity to the Accuhaler® in vitro deposition. Application of CFD modelling and material characterization of formulation raw materials enabled the modification of device and formulation critical material attributes to create an in vitro comparable device/formulation system to the reference product.KEY WORDS: aerosolization; computational fluid dynamics; device design; dry powder inhaler; in vitro comparability; in vitro performance.