I asked a similar question a few months ago, and there is a thread on it here somewhere with the title "reliability" or some such, with information that may be useful to you. Early war rotary engines had a TBO for a complete overhaul of around 10-20 hours, the mid-war types (Le Rhone, Oberursel, Clerget) of around 30-50 hours (but less, if synthetic castor oil was used), and the late war types (Bentley BR1, BR2) 40-50 hours (W.O. Bentley claimed 50 hours plus). This is compared to stationary engines that had a TBO for a complete overhaul of between 60 hours (Hispano-Suiza) to 100 hours plus (Rolls Royce Eagle), but a general average of about 80 hours. The later radials, in the immediate post-war period, about 100-150 hours (the 1922 manual for the Bristol Jupiter quotes 150 hours TBO for a complete overhaul), but then a considerably longer TBO from the mid-1920s onwards (up to 226 hours for the Jupiter VI in the mid 1920s, and 555 hours by 1931). There were, of course, several partial overhaul stages between each complete overhaul, and WWI rotaries were noted for their ease of access and maintenance in the field as compared to stationary engines (the Rolls Royce, an extreme case, had to be returned to the manufacturer for a complete overhaul at the end of its TBO interval).

Although there appears to be some relationship between reliability and TBO interval for a complete overhaul, a short TBO interval is not necessarily an accurate guide to unreliability. From the evidence that I have seen a properly maintained rotary can be as reliable as a well maintained radial or stationary engine of the same period - just with a shorter TBO interval to keep it that way.

Bletchley