This paper investigates a smart lane-change mechanism of automated vehicles with intelligent driver model on highways with multiple lanes. In particular, we propose an autonomous lane-change maneuver to improve traffic flow. The key technique of the proposed maneuver is to consider not only selfish conditions to track individual velocity tracking performance but also altruistic conditions to improve the driving performance of the succeeding vehicles by using vehicle-to-vehicle communications. To realize the conditions, this paper presents cost functions and criteria mathematically. We also demonstrate through simulation that the proposed algorithm enhances heterogeneous traffic flow efficiency while avoiding collisions and wasteful traffic congestion in dense traffic. As a result, each vehicle with the proposed maneuver can travel about 1 minute per 10 km faster than only the conventionally selfish condition on average.