An approach is provided for predicting starting points and/or ending points for traffic jams in one or more travel segments. The approach involves processing and/or facilitating a processing of probe data associated with at least one travel segment to cause, at least in part, a generation of at least one speed curve with respect to a distance dimension and a time dimension, wherein the probe data includes speed information, and wherein the at least one speed curve indicates at least one previous starting point, at least one previous ending point, or a combination thereof for one or more previous traffic jams based, at least in part, on the speed information. The approach also involves processing and/or facilitating a processing of the at least one previous starting point, the at least one previous ending point, or a combination thereof to determine at least one starting point trend curve, at least one ending point trend curve, or a combination thereof with respect to the distance dimension and the time dimension. The approach further involves determining at least one predicted evolution of at least one starting point, at least one ending point, or a combination thereof for at least one traffic jam in the at least one travel segment based, at least in part, on the at least one starting point trend curve, the at least one ending point trend curve, or a combination thereof.