The performance of a two-liquid phase biotrickling filter (TLP-BTF) packed with pall ring and pumice in the presence of silicone oil, as the organic solvent (10% v/v), was evaluated for the treatment of toluene vapor under various transient conditions. The experiments were done at EBRT of 1 min and inlet loading rates (ILRs) of 150-180 g m(-3) h(-1) and comprise of shock loading, intermittent loading for 10 h day(-1), and aeration without toluene loading during shutdown periods. The TLP-BTF removal efficiency (RE) attained steady state (RE = 66%) in less than 40 min after loading initiation in different operating strategies. The TLPBTF was subjected to a 5.5-fold increase at ILRs from 34 to 186 g m(-3) h(-1), however, RE declined only 10% within 2 h after shock loading. Nevertheless, RE decreased to 44% due to dominating the kinetic-limitation regime and silicone oil saturation. The variations of TLP-BTF performance were analyzed by using artificial neural networks (ANNs) modeling by means of casual index (CI) values. Lower discrepancies for CI-values in TLP-BTF (-8.05 > CI >-10.68) in comparison to the single-liquid phase BTF (-9.19 > CI >14.04) indicated the role of silicone oil for decreasing the negative impact of different intermittent loading types on the BTF performance. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.