Investigating the effects of in-vehicle warning strategies to drivers: A driving simulator study

• A driving simulator was used to study the impact of warning strategies on driving. • Three kinds of hazardous scenarios (bus stop, parking area, crosswalk) were tested. • High-level warnings 150 m ahead were optimal for bus stop scenarios. • Medium-level warnings 150 m ahead were optimal for roadside parking scenarios. • High-level warnings 100 m ahead were optimal for crosswalk scenarios. With the advancement of in-vehicle warning system technology, it has become feasible to develop effective warning strategies to enhance drivers’ risk perception and mitigate crash occurrence. However, the critical issues of when to deliver the warnings and what content should be included remain unresolved. The objective of the study was to investigate the effects of in-vehicle warning strategies on drivers in terms of warning information and delivery distances. An experiment using a driving simulator was conducted, during which driving behavior and eye movement data were collected from 40 participants. Nine warning strategies, including the combination of 3 levels of warning information with 3 delivery distances, were developed for the experiment. Three hazardous scenarios, including a bus stop, roadside parking area, and crosswalk, were designed in the experiment. Repeated measures multivariate analysis of variance (RM-MANOVA) tests and the EWM-TOPSIS (entropy weight method-technique for order preference by similarity to an ideal solution) model were employed to evaluate the effectiveness of these warning strategies. The results indicated that delivering high-level warning information 150 m in advance was the most effective warning strategy for the bus stop scenario. As for the roadside parking scenario, providing medium-level warning information 150 m in advance emerged as the optimal warning strategy. Regarding the crosswalk scenario, the best warning strategy was to deliver high-level warning information 100 m in advance. Moreover, providing more detailed information enhanced drivers’ performance across all scenarios. These insights can serve as a valuable reference for the design of in-vehicle warning systems.