The most efficient BE study to evaluate 3 different formulations is a 3 treatment, 3 period, 6 sequence, crossover study. A nice example of this study design can be found here: https://www.dovepress.com/cr_data/article_fulltext/s78000/78713/img/DDDT-78713-F02.jpg

If this is a pilot BE study this is the best design.However, if this is a registration study which you plan to submit to a regulatory authority, the preference per the FDA guidance is to perform two separate 2-way crossover studies (see FDA guidance: https://www.fda.gov/downloads/drugs/guidances/ucm377465.pdf.

However, the FDA will give you some leeway, as stated within the guidance:

For applicants wishing to use variations of these study designs or analysis methods (e.g., a sequential design or scaled-average BE), we recommend that you submit a complete protocol for review and comment before starting the study.

As such, if you move forward with a 3 treatment, 2-period, 6-sequence design, it would be best to consult with the FDA prior to conducting the study.

Another important guidance: https://www.fda.gov/downloads/Drugs/Guidances/ucm154838