Lane-changing manoeuvre is one of the risky manoeuvres performed by drivers either to reach the planned destination (i.e., mandatory lane-changing; MLC) or to achieve better driving conditions (i.e., discretionary lane-changing; DLC). Essentially both lane-changing types require the driver to acquire surrounding traffic information for efficient and safe lane-changing decisions. However, this does not discount the fact that both these lane-changings are fundamentally different from each other as the urgency of lane-changing is much higher during MLC compared to DLC. Real-time driving aids in a connected environment promise to assist during the lane-changing decision-making process, but the differential effectiveness (or usefulness) of real-time driving aids in a connected environment for these two lane-changing types remains unexplored due to the novelty of a connected environment and the consequent scarcity of data. To fill this research gap, this study collected lane-changing data from 78 participants who performed MLC and DLC in the CARRS-Q Advanced Driving Simulator. Participants were asked to drive in three randomised driving conditions: baseline condition (without driving aids), real-time driving aids in a connected environment with perfect communication, and delayed driving aids in a connected environment. While surrogate measures of safety are analysed and compared using descriptive statistics, a hybrid framework of data mining and classical statistical modelling is employed to examine the usefulness of the real-time driving aids in a connected environment for two lane-changing types. We find that the crash risk associated with MLC is significantly reduced in the driving conditions providing driving aids compared to that of DLC. Results also reveal that the probability of engaging in a hard-braking event decreases for both the lane-changing types in the presence of real-time driving aids in the connected environment driving conditions, but a higher decrease in magnitude is found for MLC. Age and gender-related differential impact have been observed where young and male drivers have a higher possibility of engaging in a hard-braking event when driving without driving aids, but the presence of real-time driving aids reduces such risk. This study concludes that the usefulness (or effectiveness) of the real-time driving aids in a connected environment is a function of the urgency of a task, which is evidently higher during MLC, thus providing the maximum advantage during MLC.

变道操纵是驾驶员为了到达计划的目的地（即强制变道； MLC）或达到更好的驾驶条件（即任意变道； DLC）而进行的危险动作之一。本质上，两种换道类型都要求驾驶员获取周围的交通信息，以进行有效且安全的换道决策。但是，这并不妨碍这样的事实，即这两个车道转换在根本上是彼此不同的，因为与DLC相比，MLC期间车道转换的紧迫性要高得多。互联环境中的实时驾驶辅助系统有望在换道决策过程中提供帮助，但是由于连接环境的新颖性以及随之而来的数据匮乏，在这两种车道变换类型的连接环境中，实时驾驶辅助设备的不同有效性（或有用性）仍未得到开发。为了填补这一研究空白，本研究收集了来自在CARRS-Q高级驾驶模拟器中执行了MLC和DLC的78位参与者的换道数据。要求参与者在三种随机驾驶条件下驾驶：基准条件（无驾驶辅助设备），在互联环境中实现完美沟通的实时驾驶辅助工具以及在互联环境中的延迟驾驶辅助工具。在使用描述性统计数据分析和比较替代安全措施的同时，数据挖掘和经典统计模型的混合框架用于检查在连接环境中实时驾驶辅助工具对于两种车道变换类型的有用性。我们发现，与DLC相比，在提供驾驶辅助的驾驶条件下，与MLC相关的碰撞风险显着降低。结果还表明，在连接的环境驾驶条件下，在存在实时驾驶辅助工具的情况下，对于两种车道变更类型，发生硬刹车事件的可能性均降低，但MLC的降低幅度更大。已经观察到年龄和性别相关的差异性影响，在没有驾驶辅助工具的情况下，年轻驾驶员和男性驾驶员更有可能参与硬制动事件，但是实时驾驶辅助工具的存在降低了这种风险。