Variable Message Signs in Traffic Management: A Systematic Review of User Behavior and Future Innovations

Abstract

Effective traffic management is crucial in addressing the growing complexities of urban mobility, and variable message signs (VMSs) play a vital role in delivering real-time information to road users. Despite their widespread application, there is limited comprehensive understanding of how VMS influence user behavior and optimize traffic flow. This systematic literature review aims to address this gap by examining the effectiveness of VMS in shaping user interactions and enhancing traffic management systems. Using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) methodology, a thorough analysis of relevant studies was conducted to identify key factors influencing VMS impact, including message content and characteristics, complementary sources of information, user demographics, VMS location, and users’ reliance on these signs. Additionally, the review explores the implications of displaying non-critical information on VMS and introduces virtual dynamic message signs (VDMSs) as an innovative approach for delivering public traveler information. The study identifies several research gaps, such as the integration of VMS with vehicle-to-everything (V2X) technologies, navigation systems, the need for validation in real-world scenarios, and understanding behavioral responses to non-critical information on VMS. This review highlights the importance of optimizing VMS for improved user engagement and traffic management, providing valuable insights and directions for future research in this evolving field.

1. Introduction

Variable message signs (VMSs) are electronic display panels that provide real-time information, playing a key role in road traffic management by influencing driver decision-making and behavior. These panels have evolved from traditional static signs into dynamic displays that can adapt to diverse informational needs [1]. The dynamic nature of VMS allows them to display real-time information using both text and graphics. They are used in various scenarios, such as communicating traffic conditions, road closures, emergencies, public transportation details, and parking availability.

The effectiveness of VMS largely depends on how well the content and format of the messages align with drivers’ expectations and informational needs. Key constructs that determine the perceived quality of VMS include drivers’ attitudes towards the formats and contents of the messages, decision-making processes, and the overall effectiveness of VMS messages [2,3]. Several studies have used drivers’ route change decisions as a proxy for evaluating VMS effectiveness, suggesting that this behavior reflects the perceived quality of the information provided [4,5]. However, relying solely on route change behavior overlooks other external factors that can influence drivers’ perceptions, such as the perceived accuracy, reliability, and relevance of the information displayed [6].

Research indicates that drivers’ perceptions of VMS are significantly affected by their attitudes towards specific attributes of the signs. For example, content relevance, clarity, and message formatting are all critical factors that can impact the perceived quality of service offered by VMS [2]. Studies have found that drivers are more likely to respond positively to VMS when the information is perceived as accurate and reliable, highlighting the importance of aligning message content with drivers’ needs [7]. The authors also suggest that the effectiveness of VMS messages plays a crucial role in driver acceptance of traffic information, emphasizing the need for clear and concise communication [7]. Despite their advantages, VMS are often underutilized, as users do not always find the information presented to be useful or relevant [6]. This underutilization points to a significant gap in understanding how VMS can effectively influence user behavior and improve traffic management.

The motivation for this study stems from the critical need to optimize the use of VMS in traffic management systems. Although VMS have the potential to enhance traffic flow and safety, the effectiveness of these signs is limited by how well the information aligns with user needs and preferences [6]. Understanding the factors that influence how users perceive and interact with VMS is essential to maximizing their impact on traffic management and citizen engagement. Currently, there is limited comprehensive research that explores these relationships in depth, highlighting the necessity for this study.

The objective of this research is to gather insights into how VMS influence user behavior, particularly within the context of traffic management and public engagement. This study aims to explore the interactions between VMS and users to understand how they perceive and react to the information displayed. By analyzing various studies, this research identifies key factors that influence user decision-making processes and assesses whether VMS information meets user expectations and needs.

This study categorizes findings into three main areas: the influence of VMS on user behavior, the implications of displaying non-critical information on VMS, and the introduction of the new concept of virtual dynamic message signs (VDMSs). Additionally, it identifies research gaps and suggests future directions, such as integrating VMS with navigation systems, validating findings in real-world scenarios, and understanding user responses to non-critical information on VMS. Ultimately, the goal is to contribute to a deeper understanding of VMS effectiveness and its implications, thereby optimizing their role in traffic management systems.

The structure of this article is as follows: Section 1 introduces the context and background of the research. Section 2 describes the methodology employed in the systematic literature review. Section 3 presents the findings from the review. Section 4 discusses the key themes identified in the studies, along with research gaps and future directions. Finally, Section 5 concludes the article by summarizing the main findings and contributions of the study.

2. Research Method

This systematic literature review (SLR) followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement guidelines and checklist [8], a well-established and rigorous approach commonly used in systematic reviews, being the related checklist presented in Appendix A. It provides a transparent framework for conducting literature reviews, ensuring a methodical search process, and ensuring rigorous reporting of research findings [9]. While there are various methodologies for conducting literature reviews, such as meta-analyses, scoping reviews, and narrative reviews, the SLR was chosen for its structured, transparent, and reproducible nature, which ensures a comprehensive and unbiased synthesis of the existing literature. Unlike meta-analyses, which focus primarily on quantitatively aggregating statistical findings, the SLR is suitable for integrating qualitative insights and conceptual frameworks, making it ideal for a diverse and multi-disciplinary topic like human–computer interaction (HCI) and user-centered design [10]. Similarly, unlike scoping reviews, which aim to map the breadth of a field, SLRs focus on critically appraising and synthesizing relevant studies in depth [11]. This structured approach helps avoid bias and ensures reproducibility, which is essential for the study of VMS and user interaction. Other approaches, such as narrative reviews, were not considered appropriate due to their less structured nature, which could introduce bias and limit reproducibility [12].

The research was conducted in two databases, namely SCOPUS and Web of Science, selected due to their relevance and coverage of articles in fields related to the research. These two databases were chosen because of their comprehensive coverage of peer-reviewed research, rigorous indexing processes, and detailed citation data, which are highly relevant to our study on VMS and user-centric approaches. SCOPUS and Web of Science are widely recognized for their reliability and advanced filtering options, which ensure that only high-quality, peer-reviewed research is included. Other databases, such as Google Scholar, offer broader coverage, including the gray literature, conference papers, and theses; however, it lacks the advanced filtering and consistent indexing available in SCOPUS and Web of Science. This can lead to the inclusion of lower-quality or non-peer-reviewed sources, which may introduce bias into the review [13]. Moreover, the gray literature was not included in this review because of concerns over inconsistent quality and the lack of peer review, which could compromise the rigor of the SLR [14].

The main goal of the research query was to gather studies that related VMS with a user-centric approach, focusing on specific mobility segments such as drivers, pedestrians, cyclists, and public transport passengers. The objective is to gather insights on how users are influenced by VMS in their decision-making processes and whether the information presented aligns with their needs and preferences.

Therefore, a single query was carefully designed, comprising three primary components following the PCC—population, concept, and context—framework. The first part centers on the concept of “variable message signs”, encompassing variations such as “dynamic message sign”, “road display”, and “traffic display”. The second part is dedicated to the people under study, with terms like “driver”, “pedestrian”, “cyclist”, “passenger”, and “traveller”. The third part revolves around “human-computer interaction”, incorporating terms like “user-cent*”, “usability”, “user interaction”, “user experience”, “user interface”, “interaction design”, “nudg*”, “influence*”, and “behavior”/“behaviour”.

The final query was defined after several preliminary experiments performed using different combinations of topic-related terms and Boolean operators, taking into account the objective of this review, and is presented below:

((“variable message sign” OR “dynamic message sign” OR “road display” OR “traffic display”) AND (driver OR pedestrian OR cyclist OR passenger OR traveller) AND (“human-computer interaction” OR “user-cent*” OR usability OR “user interaction” OR “user experience” OR “user interface” OR “interaction design” OR ((nudg* OR influence*) AND (behaviour OR behavior))))

This query returned a total of 96 results, with no filters applied during the initial search. Figure 1 shows the number of results obtained from each database, as well as the process followed to determine the final number of articles included in the literature review. First, duplicate results were removed. After that, the titles and abstracts of each article were briefly analyzed, and those considered irrelevant to this study were excluded. Following this, articles for which full-text access was not available, despite attempts to reach out to authors or access them through institutional resources, were eliminated. Finally, articles deemed unsuitable for the study after a full-text review were excluded. In addition to the database search, the snowballing method was applied to identify further relevant studies by examining the reference lists of the included articles. This process resulted in the addition of one more article to the dataset, enriching the overall scope of the review. In the end, a total of 18 articles were included in the review, which sample size is consistent with other systematic reviews in the field of transportation [15,16,17,18] that analyze 11, 13, 18, and 19 articles, respectively. The inclusion and exclusion criteria for the articles are presented in

Table 1. Eligibility criteria for the selection of the articles.

Inclusion Criteria

Exclusion Criteria

Articles that address variable message signs (VMSs) in the context of traffic management and/or citizen engagement; Articles with a user-centric approach, exploring how citizens perceive and interact with the information presented on VMS; Articles that discuss strategies, technologies, or findings related to engaging citizens through VMS; Articles that involve the development of prototypes for VMS and their evaluation with user feedback; Articles that explore other types of information that are suitable for presentation on VMS.

Articles that do not directly relate to the enhancement of VMS in the context of traffic management and/or citizen engagement; Articles that describe technical aspects of VMS (for example, software architecture, configurability, etc.); Articles that present models for optimizing traffic management without a clear focus on VMS enhancement.

.

3. Results

After the selection process, a total of 18 studies were considered for this study, with publication dates ranging from 2008 to 2022 and encompassing nine different countries (see Figure 2).

Table 2. Overview of the articles included in the literature review.

References and Country	Study Objective	Methodology and Sample	Main Findings	Limitations and Future Work
[19] USA	Analyze driver compliance with dynamic message sign (DMS) route guidance and factors influencing route choice decisions.	A hybrid approach using a driving simulator (DS) and a stated preference (SP) survey, with over 100 participants in a 400 km2 network near Baltimore.	- Key factors: past DMS exposure, travel time savings, DMS reliability, and learning from experience. - Divergence between DS experiment decisions and SP survey responses. - Higher compliance with DMS in DS experiments compared to SP scenarios.	- Limitation: Lack of personalized navigation systems like GPS may affect generalizability. - Future work: Include navigation systems to explore their impact on driver compliance.
[20] USA	Investigate the impact of different DMS types (including color-coded and alphanumeric) on driver behavior, focusing on route diversion, choice, and compliance.	A high-fidelity driving simulator at Morgan State University with six virtual scenarios was used. Sixty-five participants from diverse backgrounds experienced various DMS messages. Data were collected from pre- and post-simulation surveys and driving behavior.	- Route diversion: DMSs with delay and lane closure information were most effective. - Route choice: color-coded DMSs and “avoid route” advice had a strong impact. - DMS compliance: color-coded DMSs and crash-related advice increased compliance; time-distance alternate routes reduced compliance.	- Limitation: No navigation system was included. - Future work: Study the interaction between GPS and DMS and explore standards for autonomous vehicle integration.
[21] Spain	Evaluate two PARC (Parking Access and Revenue Control) systems—level 4 and level 2—and their impact on user search times and parking choices.	Simulation of a parking facility using a 3D model, comparing level 4 (censoring info) and level 2 (zoning with vehicle detection) PARC systems.	- Level 4 PARC is effective in small facilities but has little impact on search times. - Level 2 PARC reduces search times by 16.2%. - Availability information helps users make better parking choices. - Manipulating info in level 2 systems reduces search times.	- Limitation: impact of informing users about specific level shortages needs further study. - Future work: field experiments or real-world implementations to validate results.
[22] Italy	Investigate how different VMS message displays (uppercase/lowercase letters and familiar/less familiar pictograms) affect reading time and information perception, focusing on drivers of different ages and speeds.	A simulated driving scene in a controlled environment at the University of Cagliari. Messages were displayed at speeds of 50 km/h and 80 km/h. Eye-tracking technology was used to measure reading times for 30 drivers, categorized into young, middle-aged, and elderly groups.	- Uppercase vs. lowercase letters did not significantly impact reading times. - Unfamiliar pictograms hindered message perception. - Elderly drivers had more difficulty perceiving messages compared to younger groups.	- Limitation: the simulation environment may not fully replicate real-world conditions. - Future work: use a physical driving simulator for more realistic experiments.
[23] China	Investigate urban freeway users’ diversion responses to a dual variable message sign (D-VMS) showing travel times for both freeway and local streets and examine factors influencing route choice behavior.	An on-site stated preference survey in Shanghai with 140 drivers. Participants were surveyed based on gender, age, driving experience, frequency of freeway use, and driver type, using scenario-based surveys with D-VMS displaying travel times and causes of delays.	- D-VMS significantly impacts diversion decisions. - Key factors: travel time savings, years of driving experience, traffic lights on local streets, frequency of freeway use, age, and driver type. - Panel models provide more robust statistical results than cross-sectional models.	- Limitation: small sample size due to road changes. - Future work: focus on advanced traveler information systems and improving survey and model design in urban transportation.
[24] China	Identify factors influencing drivers’ route choice response to travel time information provided by variable message signs (VMSs) on arterial roads.	A stated preference survey was conducted in Shanghai with 228 drivers, resulting in 1512 choice observations from 189 drivers. Generalized Estimating Equations (GEEs) were used to model driver responses, comparing four different correlation structures.	- Key factors: driving experience, expressway delay, cause of delay (accident), and the number of traffic lights on local streets. - Differences in behavior among drivers of employer-provided cars, taxis, and private cars. - The GEEs method was effective, with the exchangeable structure being the most suitable.	- Limitation: study limited to Shanghai. - Future work: cross-country comparisons and collecting revealed preference data after implementing enhanced VMS services.
[25] Netherlands	Study the impact of using VMS for both traffic management and displaying non-traffic messages, such as ads or mottos.	Experimental design with 32 participants divided into two groups—control (no ads) and experimental (ads)—using a driving simulator.	- Non-traffic messages on VMS do not negatively affect traffic management if ergonomic guidelines are followed. - Compliance with critical traffic instructions was higher in the group exposed to ads. - Conscious attention was not required for compliance. - Familiarity with VMS through repeated exposure positively influenced driver behavior.	- Limitation: ads were designed to minimize distraction, which may not fully represent typical commercial ads. - Future work: investigate the impact of non-ergonomic messages and real-world distraction thresholds.
[26] Netherlands	Study how drivers respond to different sources of traffic information, such as variable message signs (VMSs) and in-car navigation devices.	A driving simulator was used to simulate route choice scenarios with varying traffic information sources (VMS and in-car navigation). Participants rated their trust in the information after each trial, and a Bayesian model was used to analyze route-switching likelihood. Twenty-four participants took part in the study.	- Trust in traffic information was higher when provided by two sources versus one. - High compliance with the preferred source, mainly VMS. - Participants were more likely to switch routes when the alternative route was shorter. - Age, gender, and yearly mileage did not significantly affect route-switching propensity.	- Limitation: the sample consisted of experienced drivers, limiting generalizability. - Future work: explore how other traffic information sources interact and influence behavior towards VMS.
[27] Germany	Investigate the impact of individual and collective on-trip traffic management systems on traffic flow in urban networks and how traffic information design influences route choice behavior.	Three systems were considered: radio traffic service, variable traffic signs, and navigation systems. Evaluation methods included ANPR, GPS-logging, interviews, and a driving simulator. The study tracked 300 individuals in Munich over 8 weeks.	- Route choice behavior was mainly influenced by travel time. - Drivers preferred the main route unless there was a significant travel time increase. - Radio traffic services and navigation systems significantly influenced individual decisions. - The driving simulator highlighted the role of familiarity and subjective motivations in route choice.	- Limitation: difficulty studying individual behavior in detail. - Future work: investigate the impact of emerging systems and reliability of route choices.
[1] Austria	Design a dynamic traffic display that effectively conveys information to drivers and promotes behavioral change, exploring various design categories based on four dimensions.	The study used a conceptual design based on four dimensions: emotionality, complexity, information content, and familiarity. An online survey was conducted to eliminate poor designs, followed by a lab evaluation with 12 participants sitting in front of a simulated car cockpit.	- “Icon oriented”, “schedule oriented”, and “traffic sign oriented” designs were rated higher than “emotion oriented” signs. - Icon-oriented signs were clear favorites, being unique, memorable, and easily interpreted.	- Limitation: subjectivity in design evaluation was noted. - Future work: further exploration of designs that support long-term behavioral change, with a focus on clarity for individuals with foreign mother tongues, is needed.
[28] USA	Propose and evaluate the concept of virtual dynamic message signs (VDMSs) as a future method for delivering public traveler information, comparing it to traditional dynamic message signs (DMS) in terms of user acceptance, message comprehension, and driver distraction.	A smartphone-based VDMS prototype was developed. The study included a user experience survey with 21 participants from Northern Virginia who used the VDMS app for two weeks and a driving simulator study with 42 participants to assess message comprehension and driver distraction in various driving scenarios.	- Positive user experience with VDMS, with high ratings for usefulness and safety. - VDMS was seen as safer and more comfortable than traditional DMS. - The driving simulator study showed VDMS performed better than DMS, especially under complex driving conditions. - VDMS was as effective as DMS under simple conditions but significantly better in complex scenarios.	- Limitation: on-road assessments are needed to confirm simulator findings. - Future work: explore additional factors like gender and education level and evaluate the effectiveness of improved auditory VDMS messages in real-world settings.
[29] USA	Assess the relationship between traffic crashes and the frequency of safety messages displayed on dynamic message signs (DMSs) on Michigan freeways, examining if these messages influence driver behavior and crash risk.	Data from 202 fixed DMS on Michigan freeways (2014–2018) were integrated with traffic volume, roadway geometry, and crash data. Random parameters negative binomial models were used to analyze total, speeding-related, and nighttime crashes.	- No significant difference in total crashes related to message frequency. - Marginal, statistically insignificant declines in nighttime crashes with more impaired driving messages. - Significant reduction in speeding-related crashes near DMS with frequent messages about speeding or tailgating.	- Limitation: uncertainty about how many drivers read or retain messages and the influence of urban DMS density. - Future work: field research to explore the impact of safety messages on individual driver behavior.
[30] Germany	Evaluate how information from variable message signs (VMSs), broadcast traffic updates, and other factors influence drivers’ route choices, focusing on the impact of VMS recommendations and broadcasted traffic information.	The study used floating car data (FCD) from mobile phones to analyze route choices through maximum likelihood estimations, logit models, and utility functions. Two cases were studied: the motorway quadrangle and Stuttgart.	Motorway quadrangle: - Drivers respond to congestion proximity in broadcast traffic updates. - VMS recommendations diverted 30% of through-traffic to alternative routes. Stuttgart: - Results were statistically insignificant, with only 3% to 17% of drivers accepting VMS recommendations.	- Limitation: Infrequent VMS activation in Stuttgart led to inconclusive results. - Future work: investigate external factors (e.g., weather or events) and how drivers adapt their route choices based on dynamic variables beyond traffic congestion.
[31] United Kingdom	Explore the influence of dynamic information on VMS on driving behavior and decision-making, focusing on factors like familiarity, information wording, and context.	Two methods were used: a scenario-based approach and a medium-fidelity driving simulator. Data were collected from 82 UK drivers (aged 21–65) through scenarios, questionnaires, and driving tasks.	- Repetition of information increases trust and encourages drivers to slow down. - Additional details about traffic conditions (e.g., delay length) affect driver behavior. - Specific information, like mentioning an “accident”, prompts more cautious driving. - Context and prior experience influence decision-making.	- Limitation: caution is needed when implementing complex signs in real-world contexts. - Future work: explore the role of new information sources (e.g., sensor-based tech and social media) for more effective traffic information design.
[32] China	Investigate the effects of VMS on driving behaviors in urban areas, focusing on factors such as age, driving experience, familiarity with the road network, and attention to VMS. The study also aimed to develop a logistic model to analyze the influence of VMS on driving behavior.	A questionnaire survey was conducted with 578 samples from car repair shops and parking lots in Yangzhou, China, with 402 valid responses. A logistic model was developed based on survey data and real-world traffic data from video cameras, along with correlation analysis.	- Main factors affecting driving behavior: age, driving experience, road familiarity, and attention to VMS. - Drivers aged 36–55 and those with 6–10 years of driving experience were more likely to reduce speed to read VMS. - Drivers unfamiliar with the road network also tended to reduce speed to read VMS. - The logistic model confirmed that VMS significantly affects driving behavior in urban areas.	- Limitation: sample bias from data collected in Yangzhou, China, may limit generalizability. - Future work: further research on the optimal design of VMS messages, including size, color, and layout, is recommended.
[4] Greece	Investigate factors influencing driver response to VMS in Athens, focusing on route diversion due to incidents. The study examines the impact of message characteristics, trip characteristics, and driver characteristics on diversion behavior.	A stated preference questionnaire survey was conducted with 120 Athenian drivers familiar with VMS. The survey covered traffic information preferences, trip details, and incident scenarios. A random-effect ordered probit model was used for analysis, with data collected through on-site interviews.	- Message characteristics (incident type, impact, and alternative route) significantly influenced diversion decisions. - Accidents were the most influential incident type, followed by demonstrations and roadworks, with congestion being the least influential. - Qualitative descriptions of impact (e.g., ‘long delays’) had a stronger effect than quantitative descriptions. - Drivers were more likely to divert when an alternative route was suggested via VMS. - Trip and driver characteristics (e.g., transport mode, VMS experience, attitudes towards VMS) also influenced decisions.	- Limitation: the findings are based on Greek drivers and may not generalize to other urban areas. - Future work: explore diversion behavior in other driver populations (e.g., professional drivers) and tailor VMS information to specific driver groups. The study also recommends improving VMS reliability by withdrawing messages once incidents end.
[33] China	Test the effectiveness of VMS on driving behavior, focusing on how VMS location and information format influence route choice, speed control, and lane-changing behaviors.	Participants drove three simulated scenarios using a high-fidelity driving simulator. Two parts of the network were designed to test the impact of VMS location and information format. Fifty-seven subjects participated, with data collected on route choice, speed, lane changes, and other behaviors.	- VMS location, information format, and driver characteristics (age, gender) significantly influenced driving behavior. - VMS placed 150–200 m upstream from the diverging point was optimal for safety and traffic flow. - Graphic information on VMS was more effective than text-only formats. - Older drivers were less likely to change routes, while male and professional drivers were more influenced by VMS.	- Limitation: the study did not account for socioeconomic factors or confidence in VMS information. - Future work: explore more factors, including VMS content, to better understand their combined impact on driving behavior.
[34] China	Analyze the causal relationships among drivers’ attributes, learning behavior, reliance on VMS, and route choice behavior in the context of road networks in Beijing.	Data were gathered from the “Inducing Effect Survey of VMS in Beijing”, with 311 valid responses from 350 distributed questionnaires. The study focused on traffic information from 228 VMS devices across major trip generators in Beijing. A structural equation model (SEM) was used for analysis.	- Drivers’ learning from VMS significantly impacts their reliance on VMS. - Learning behavior is a key factor in route choice decisions, influenced primarily by driving years and age. - Drivers who trust and rely on VMS are more likely to follow alternative routes suggested by the signs.	- Limitation: the data used in the study had limitations, and more observed variables should be considered. - Future work: further research should include more observed variables to strengthen the findings.

serves as the primary framework for the literature review and comprises five columns. The first column outlines the authors, the publication year of the article, and the country where the study was conducted; the second column refers to the study objective; the third column outlines the methodology employed in the study and the sample used; the fourth column summarizes the main findings; and the fifth column refers to the limitations of the study and future work. The next section presents a critical analysis of the selected articles.

Summarizing the findings from Table 2, it is possible to observe that the effectiveness of VMS in influencing route choice is significantly impacted by the content and presentation of the messages. Studies consistently demonstrate that detailed information about delays, alternative routes, or specific instructions, such as using color-coded versus alphanumeric formats, enhances compliance and influences route diversion decisions [9,10,13,20]. The format and clarity of the messages are crucial; familiar pictograms or simple graphics can improve perception, while overly complex visuals can be counterproductive [12,24]. Additionally, the effectiveness of VMS is amplified when used alongside other traffic information sources, such as in-car navigation systems, radio traffic services, or mobile applications. Drivers tend to trust VMS information more when it is corroborated by additional sources, which enhances the likelihood of compliance with the suggested routes [16,17].

Furthermore, VMS impacts driver behavior in various ways, often influenced by demographic factors such as age, driving experience, and familiarity with the road network. For instance, older drivers and those less familiar with the routes are more likely to reduce speed to read VMS messages, indicating a need for targeted design considerations that cater to these groups [12,22]. In contrast, professional drivers tend to be more compliant with VMS recommendations, suggesting that those with more driving experience have a higher reliance on VMS [24,25]. Moreover, the use of VMS for non-critical information, such as advertisements or general safety messages, does not necessarily detract from their primary traffic management function. In some cases, exposure to non-critical messages has been found to increase driver engagement and compliance with subsequent critical instructions, potentially due to increased familiarity with the VMS system [15].

However, the design and ergonomics of such messages need careful management to avoid distracting drivers.

Among the innovative applications of VMS explored in the literature is the concept of virtual dynamic message signs (VDMSs), which extend the functionality of traditional VMS by delivering real-time, personalized information directly to drivers via in-vehicle devices. VDMS has shown promise in improving message comprehension and reducing driver distraction, especially under complex driving conditions, suggesting a potential future shift towards more integrated, personalized traffic management solutions [18]. Additionally, a significant research gap identified in the literature is the integration of VMS with advanced navigation systems and the need for validation of these findings in real-world scenarios. While driving simulators provide controlled environments to study driver behavior, real-world studies are essential to confirm these findings and adapt VMS applications to evolving traffic conditions and user expectations [9,10,16].

4. Discussion

VMS have become essential components of modern traffic management systems, providing real-time information to guide drivers and optimize traffic flow. There are various types of VMS, each serving specific purposes, such as displaying traffic conditions, emergencies, public transportation details, and parking availability.

This section examines studies conducted in various countries, identifying three main categories of findings. The first category focuses on the influence of VMS on user behavior, encompassing five subcategories: “message content and characteristics”, “complementary sources”, “demographics”, “VMS location”, and “reliance on VMS”. These subcategories explore specific factors that affect how VMS influences user decisions.

The second category explores the implications of displaying non-critical information on VMS, including its effects on user attention and decision-making. The third category introduces the concept of virtual dynamic message signs (VDMSs), which aims to enhance traditional VMS through innovative technologies.

Organizing the findings into these categories provides a comprehensive understanding of VMS utilization and its impact on user behavior, offering valuable insights to guide future research in transportation management.

4.1. VMS Influence on User Behavior

The analysis of the articles revealed several common factors that influence user behavior in response to VMS. These factors can be categorized into five main topics:

i.

Message Content and Characteristics

The content and characteristics of the messages displayed on VMS significantly affect user responses. In terms of content, details such as the cause and length of delays, alternative routes, and travel time savings play a crucial role in driver decision-making [4,5,6,7,8]. Additionally, the number of traffic lights on local streets is also a significant factor influencing driver diversion, as noted in [6,7].

Regarding message characteristics, qualitative descriptions, such as “long delays”, have a stronger influence on diversion probabilities compared to quantitative descriptions [9]. The effectiveness of VMS messages is further enhanced by graphical formats; for instance, graphical information is more effective than text-only formats [10], and color-coded VMS outperform alphanumeric ones [5]. Interestingly, the use of uppercase versus lowercase letters does not significantly impact reading times [11].

Design is also a critical factor influencing user behavior. According to [1], various VMS prototypes were created, exploring different design categories based on dimensions such as emotionality, complexity, information content, and familiarity. The four design categories include: (i) icon-oriented signs: These use recognizable symbols and emphasize simplicity and ease of understanding; (ii) schedule-oriented signs: Inspired by VMS, these provide real-time, detailed information delivery; (iii) traffic-oriented signs: These align with traditional traffic sign conventions to increase familiarity and pattern recognition; and (iv) emotion-oriented signs: These aim to evoke emotional responses through familiar characters, logos, or by intentionally generating negative emotions to influence perceptions and behavior. The study found that “icon-oriented”, “schedule-oriented”, and “traffic sign-oriented” designs were rated higher than “emotion-oriented” designs. Among these, “icon-oriented” signs were the subjective favorites due to their uniqueness, memorability, and ease of interpretation.

ii.

Complementary Sources

Complementary sources, such as radio traffic services, navigation systems, and mobile apps, play a significant role in influencing route choices. According to [12], drivers show a higher level of trust in traffic information when it is corroborated by multiple sources, such as VMS combined with in-car navigation devices. Another study [13] found that route choice behavior is primarily driven by travel time considerations, leading drivers to rely on a combination of information from radio traffic services, navigation systems, and VMS to select the fastest route. Additionally, the authors [14] suggest that drivers take the proximity of congestion into account when interpreting broadcast traffic information; the closer the congestion, the greater its influence on route decisions. Notably, 30% of drivers who typically follow their standard route were willing to switch to the route recommended by VMS when congestion was nearby.

iii.

Demographics

User responses to VMS are also affected by demographic factors such as age, gender, education level, and occupation. Several studies have identified age and driving experience as key influences on driver behavior [6,7,15,16]. Drivers aged 36–55 and those with 6–10 years of driving experience tend to read VMS at slower speeds compared to younger drivers or those with more extensive driving experience [15]. Additionally, elderly drivers often have more difficulty perceiving VMS messages, which can result in speed reductions [11]. Professional drivers are generally more likely to follow the information displayed on VMS, while older drivers are typically less inclined to change routes based on VMS guidance [16].

iv.

VMS Location

The location of VMS is crucial for achieving the desired outcomes, highlighting the importance of identifying optimal placement for these panels. The study by [16] examines how VMS location affects driving behaviors, specifically route choice, speed control, and lane changing. Using a driving simulator, the research investigates the impact of VMS messages at various distances from route-diverging points (ranging from 0 to 400 m). The findings suggest that if VMS are positioned too far from the diverging point, drivers are less likely to follow the guidance provided. Conversely, if VMS are placed too close, drivers may attempt to change routes too late, compromising safety. The study recommends an optimal VMS location of between 150 and 200 m upstream of the diverging point, balancing traffic safety and operational effectiveness.

v.

Reliance on VMS

Reliance on the information presented by VMS is one of the most significant factors influencing user behavior. According to [4], factors such as prior exposure to VMS, the perceived reliability of the information provided, and learning from past experiences are critical in shaping driver responses. Drivers are more likely to follow VMS guidance when they are familiar with the system and have had positive experiences from adhering to the information displayed. As noted by [17], the way drivers learn from VMS directly impacts their level of reliance on these signs, with learning behavior playing a key role in route decision-making. Additionally, drivers with higher education levels tend to have stronger learning capabilities, which further influences their reliance on VMS, leading them to follow the suggested alternative routes. Another study [8] supports this understanding by indicating that repeated exposure to information increases trust, resulting in drivers being more likely to slow down when a message is repeated.

4.2. Display of Non-Critical Information on VMS

In addition to displaying critical traffic information, such as messages about accidents, road closures, detours, severe weather conditions, and alternative routes that directly impact driver safety and navigation, VMS can also display non-critical information. This non-critical category includes messages that are informative but may not have an immediate effect on traffic management and control, such as advertisements, public transportation schedules, and parking availability.

A study conducted in the Netherlands investigated the impact of traffic-irrelevant messages on VMS. Using a driving simulator with 32 participants divided into a control group and an experimental group, the study found that displaying non-critical messages, when designed following ergonomic guidelines, did not negatively affect traffic management. Surprisingly, compliance with critical traffic instructions increased among participants exposed to advertisements. The findings also highlighted the positive influence of repeated exposure to VMS on driver behavior, enhancing user familiarity with the system and increasing compliance with critical traffic instructions [18].

Based on these results, it can be concluded that displaying non-critical information on VMS does not pose significant issues, provided that these messages adhere to established guidelines to ensure clear readability and minimize unnecessary distraction.

Another study conducted in Spain focused on assessing the impact of presenting real-time parking availability information on VMS to reduce search times within parking facilities. The study evaluated the effectiveness of two distinct systems: Level 4 and Level 2 PARC systems.

At Level 4, the PARC system uses censoring information, activating messages when occupancy levels in parking facilities exceed predefined thresholds. This system communicates to users through VMS when parking availability drops below certain levels. The study found this method to be particularly effective in smaller facilities, with minimal impact on search times [19].

In contrast, Level 2 PARC systems employ zoning strategies combined with vehicle detection systems. Sensors placed at various points within the parking facility divide it into internal zones, allowing the system to provide specific information about available spaces in each zone. This approach demonstrated a 16.2% reduction in search times, especially beneficial for larger facilities like those found in shopping centers [19].

The study concluded that providing space availability information on VMS positively impacts decision-making, resulting in decreased search times within parking facilities.

A more recent study in the USA explored the relationship between traffic crashes and the frequency of safety messages on VMS along Michigan freeways. The study aimed to determine whether displaying safety messages (e.g., “avoid cell phone use while driving”) on VMS influences driver behavior and, consequently, crash risk. Data from 202 fixed VMS, collected between 2014 and 2018, was integrated with traffic volume, roadway geometry, and crash data for downstream segments of each VMS [20].

The study found no significant differences in total crashes relative to the frequency of safety messages. However, marginal declines in nighttime crashes were observed at locations with more frequent messages related to impaired driving, and there were significantly fewer speeding-related crashes near VMS that frequently displayed messages about speeding or tailgating [20].

In conclusion, while the overall impact of safety messages on crash rates was limited, the reductions in nighttime and speeding-related crashes suggest that such messages can influence driver behavior. Enhancing user engagement through more targeted and interactive messages could further increase the effectiveness of these safety messages, contributing to improved road safety.

As mobile phone applications for traffic management continue to evolve, exploring innovative uses of VMS beyond traditional traffic-related information will be crucial to provide more engaging and useful content to users.

4.3. New Concept: Virtual Dynamic Message Signs and Vehicle-to-Everything

From a forward-looking perspective, the concept of virtual dynamic message signs (VDMSs) was introduced by the authors [21] as a potential method for disseminating traffic information to drivers. VDMS seeks to enhance the delivery of traveler information by replacing traditional text messages on VMS with detailed, in-vehicle audio messages delivered through smartphones. This approach offers scalability, personalization, and location-based delivery, addressing some of the limitations of current dynamic message signs (DMSs).

The study evaluated the effectiveness of VDMS compared to traditional DMS across various dimensions, including user acceptance, message comprehension, and driver distraction. Results from the user experience survey showed a positive attitude towards VDMS, with participants expressing high levels of satisfaction. VDMS was perceived as a safer alternative for receiving information, with users reporting greater comfort compared to traditional DMS. In driving simulator tests, VDMS consistently outperformed DMS across different information volumes and driving conditions, irrespective of drivers’ ages. Overall, VDMS demonstrated a similar effectiveness level to DMS in simpler conditions and surpassed DMS performance in more complex driving scenarios, highlighting its potential advantages in message comprehension and reducing driver distraction.

However, while VDMS represents a significant advancement, it is important to acknowledge the emergence of newer technologies, such as connected and automated vehicles. The focus should now shift towards integrating VMS, including VDMS, into the broader vehicle-to-everything (V2X) ecosystem, which encompasses communication between vehicles, infrastructure, and other road users. This integration can be achieved by directly connecting VMS with vehicles (vehicle-to-vehicle, V2V) and infrastructure (vehicle-to-infrastructure, V2I), facilitating real-time, location-specific information delivery that enhances message relevance and effectiveness. Through V2X communication, VMS can dynamically update information based on live traffic data and changing road conditions, offering personalized messages tailored to individual drivers’ needs, such as route-specific updates or parking availability.

Furthermore, this integration supports enhanced safety by providing timely alerts about road hazards or adverse weather conditions, enabling drivers to adjust their behavior accordingly. VMS can also interact with automated vehicle systems, supplying contextual information that assists in automated decision-making processes, such as speed adjustments or lane merging. By embedding VMS within the V2X network, these signs can contribute to a more connected, responsive, and intelligent traffic management environment, aligning with the latest advancements in connected and automated vehicle technologies.

4.4. Research Gaps and Future Directions

While significant advancements have been made in the application of VMS and their integration into modern traffic management systems, several research gaps remain that should be further explored. Addressing these gaps will not only enhance the understanding of VMS effectiveness but also optimize their role in the evolving landscape of connected and automated vehicle technologies.

i.

Advanced Integration with V2X Technologies

Current systems such as mobile apps and interactive maps provide real-time information on weather, road conditions, and traffic congestion, including VMS and DMS messages. However, there is still a need to explore how VMS can be more deeply integrated into the broader V2X ecosystem. Future research should focus on the interoperability of VMS with other V2X components, such as V2V and V2I communications. This integration could improve the relevance and timeliness of information delivered to drivers, allowing VMS to play a more dynamic role in the connected transportation network.

ii.

Empirical Validation in Complex, Real-World Scenarios

While driving simulator studies provide valuable insights into driver behavior in controlled environments, there is a need for empirical validation of these findings in complex, real-world scenarios. Future research should prioritize field studies that examine the effectiveness of VMS in diverse and variable conditions, such as urban, suburban, and rural settings. This would help to understand the practical implications of VMS, including the thresholds for driver distraction and the long-term impacts of repeated exposure to VMS messages. These studies are crucial for determining the real-world applicability and reliability of VMS as part of an integrated traffic management strategy.

iii.

Impact of Non-Critical Information in Modern Traffic Systems

Although some research has explored the effects of displaying non-critical information, such as advertisements, safety messages, and parking availability on VMS, the understanding of its broader impact on driver behavior remains limited. With the rise of mobile apps and other digital platforms for traffic management, future research should investigate how non-critical information can be effectively integrated into VMS without compromising safety or distracting drivers. This could include studies on the types of non-critical messages that are most beneficial, as well as strategies for balancing critical and non-critical information to maximize the overall utility of VMS.

iv.

Enhancing User Engagement Through Personalization and Interactivity

One emerging area for future research is the personalization and interactivity of VMS messages. As connected vehicle technologies continue to advance, there is an opportunity to explore how VMS can deliver more tailored, driver-specific information based on real-time data, such as individual driving habits, route preferences, or vehicle-specific requirements. Research should investigate the potential benefits of interactive VMS that allow drivers to receive personalized recommendations, contributing to a more user-centric approach to traffic management.

Addressing these research gaps will not only contribute to a better understanding of VMS effectiveness and its implications but also provide insights for optimizing its role and contribute to a more effective traffic management system.

5. Conclusions

In conclusion, this systematic literature review provides valuable insights into the influence of variable message signs (VMSs) on user behavior and their potential to enhance traffic management systems. Through an extensive analysis of 18 studies, five key factors were identified as crucial in shaping user interactions with VMSs.

Firstly, the content and characteristics of VMS messages were found to significantly impact user decision-making. Elements such as providing alternative route information and the graphical presentation of data were particularly influential in affecting user responses. Secondly, the role of complementary information sources was highlighted. Users demonstrated greater trust in VMS information when it was supported by additional data from sources like radio traffic services and navigation systems, which helped corroborate the VMS messages.

Additionally, demographic factors such as age, driving experience, and education level were shown to influence user responses to VMS. Recognizing these demographic differences is essential for tailoring VMS messages effectively to various user groups. Moreover, the strategic placement of VMS panels along roadways was identified as a critical factor for their effectiveness. Optimal positioning, considering factors such as visibility, timing, and proximity to route-diverging points, is necessary to facilitate informed decision-making by users. Lastly, users’ reliance on VMS information was a key determinant of their behavior. Factors like familiarity with the system, perceived reliability of the information, and past experiences influenced user trust and reliance on VMS guidance.

An important finding also emerged regarding the display of non-critical information on VMS, indicating that when presented in ergonomic formats, such information does not compromise traffic management and may even enhance compliance with critical instructions. Furthermore, the introduction of virtual dynamic message signs (VDMSs) offers a forward-looking perspective on public traveler information delivery, providing a more personalized and location-based approach to conveying information traditionally displayed on VMS.

The review also identified several research gaps that highlight the need for future exploration, particularly in integrating VMS with navigation systems, validating findings in real-world scenarios, and understanding the behavioral response to non-critical information. Additionally, there is a significant opportunity to explore the integration of VMS within the V2X communication ecosystem. By connecting VMS with V2X technologies, VMS can become part of a more interconnected, responsive, and intelligent traffic management environment, enhancing real-time communication between vehicles, infrastructure, and road users. Addressing these gaps will enable the development of more effective and user-centric traffic management strategies, ultimately contributing to more efficient and safer mobility solutions.