Different joint types and geometries have been considered to measure the fracture toughness ( Gc) of adhesives. However, results show that, sometimes, the obtained Gc values for the same adhesives are not similar in different reports. Several factors including the joints geometry, material properties of the substrate, and also the test conditions influence the obtained results. This study is conducted to find a logical connection between these factors and to find their influences on the obtained mode I fracture toughness ( GIc) to understand which parameters are the most influential and which ones are less significant. To this aim, 115 values of different GIcalready reported in the literature were collected and examined regarding the geometrical and material parameters. To find the rational relationship between the effective parameters and the reported GIcvalues, a method based on the artificial neural network technique was employed. The results revealed that whilst the fracture energy of aluminum adhesive joints is more influenced by geometrical parameters including the joint type and substrate thickness, the steel adhesive joints are more sensitive to the adhesive properties and bondline thickness. With this study, it is possible to design an optimum test by minimizing the effects of variables that cause errors in obtaining fracture toughness and also to estimate GIc of adhesives by using the developed model. On the other hand, it is also possible to design real structures in which the fracture toughness of the adhesive reaches its maximum value.