Efficient utilization of the transport infrastructure requires appropriate control and management strategies. Macroscopic traffic stream models are essential for developing such strategies. Modelling any physical system involves the identification of certain system variables which characterize its performances. Barring a few exceptions including the concept of area occupancy, progress in this direction is negligible. Present study attempts to characterize the heterogeneous, no lane-disciplined traffic stream by appropriately defining various measures of the traffic states. The authors have modified the Edie's generalized definitions by incorporating an additional dimension of the space. The proposed definitions are capable of capturing the vehicle heterogeneity as well as the no lane-disciplined driving. Advantages of the proposed definitions were tested over an hypothetical traffic stream as well as using the empirical data and found that the proposed definitions are useful in characterizing the heterogeneous traffic stream. © 2018 The Author(s).