Good logistical management optimizes offshore air transport activities, making them more efficient and reducing costs for the contractor.A series of strategic decisions, such as hiring helicopters and investments in airport infrastructure are dependent on forecasting passenger demand. The present work consisted of analyzing the demand for Petrobras offshore air transport to the State of Rio de Janeiro, based on the main theories of hierarchical time series, with the objective of identifying which of these is more suitable for a twelve-month steps ahead forecast. The strategies of single-level approach (bottom-up and top-down), optimal reconciliation (ordinary least squares and weighted least squares) and trace minimization (sample covariance and shrink estimator) were analyzed, all using exponential smoothing as the basic forecasting method. Data from 2014 to 2019 were gathered for all aerodromes used by Petrobras in the State of Rio de Janeiro: Farol de São Tomé, Campos dos Goytacazes, Macaé, Cabo Frio and Jacarepaguá. The results were evaluated with three different metrics of accuracy: RMSE (Root Mean Square Error), MAPE (Mean Absolute Percentage Error) and MASE (Mean Absolute Scaled Error), applied to the two existing levels of aggregation. The results were ranked for each technique, in the three metrics mentioned above, and then consolidated using a simple arithmetic mean. The overall results indicated that sample covariance trace minimization method provided the most accurate results.