Evaluating the Safety and Cost-Effectiveness of Shoulder Rumble Strips and Road Lighting on Freeways in Saudi Arabia

Abstract

This study examines the safety and cost-effectiveness of implementing shoulder rumble strips (SRS) and road lighting on Saudi Arabian freeways, providing insights into their roles in fostering sustainable transport systems. By leveraging the Highway Safety Manual (HSM) framework, this research develops localized Crash Modification Factors (CMFs) for these interventions, ensuring evidence-based and context-specific evaluations. Data were collected for two periods—pre-pandemic (2017–2019) and post-pandemic (2021–2022). For each period, we obtained traffic crash records from the Saudi Highway Patrol database, traffic volume data from the Ministry of Transport and Logistic Services’ automated count stations, and roadway characteristics and pavement-condition metrics from the National Road Safety Center. The findings reveal that SRS reduces fatal and injury run-off-road crashes by 52.7% (CMF = 0.473) with a benefit–cost ratio of 14.12, highlighting their high cost-effectiveness. Road lighting, focused on nighttime crash reduction, decreases such crashes by 24% (CMF = 0.760), with a benefit–cost ratio of 1.25, although the adoption of solar-powered lighting systems offers potential for greater sustainability gains and a higher benefit–cost ratio. These interventions align with global sustainability goals by enhancing road safety, reducing the socio-economic burden of crashes, and promoting the integration of green technologies. This study not only provides actionable insights for achieving KSA Vision 2030’s target of improved road safety but also demonstrates how engineering solutions can be harmonized with sustainability objectives to advance equitable, efficient, and environmentally responsible transportation systems.

1. Introduction

Road safety remains a critical global challenge, with the World Health Organization reporting about 1.2 million fatalities annually from traffic crashes, making it a leading cause of death worldwide [1]. Efforts to address this issue have led to international initiatives such as Vision Zero, which aims to eliminate traffic fatalities through a comprehensive strategy that integrates engineering, enforcement, education, and policy reforms. Technological advancements, including connected vehicles, adaptive lighting, and automated safety systems, have further showcased their potential to significantly improve safety outcomes. However, the success of these measures is heavily influenced by context-specific factors, such as infrastructure quality, driver behavior, and environmental conditions. To effectively address these challenges, it is essential to implement localized interventions that are grounded in robust, region-specific data. This need is particularly pronounced in high-income countries like Saudi Arabia, which face unique road safety dynamics and opportunities for impactful improvements.

In the Kingdom of Saudi Arabia (KSA), traffic crashes are a leading cause of mortality among young people and children in the Kingdom. According to 2016 data, the KSA had the highest rate of traffic-related fatalities among high-income countries, with a rate of 28.8 fatalities per 100,000 capita [2]. In response, the Kingdom has implemented various measures to improve road safety with the aim of reducing fatalities to 8 per 100,000 capita by 2030. As of 2023, the fatality rate has decreased to 13.20 fatalities per 100,000 capita [3], which is a result of the collaborative efforts between government traffic safety-related authorities. They have implemented several initiatives related to enforcement, such as the installation of speed cameras, red-light running cameras, seat-belt cameras, and phone usage cameras along the Kingdom’s roads. However, engineering interventions are critical for further progress to achieve the goal by 2030. These interventions align with the concept of “forgiving roads,” which aim to minimize the consequences of driver errors through safe and sustainable road designs. Enhancing infrastructure with sustainable technologies, such as solar-powered lighting and long-lasting road treatments, can reduce environmental impact while improving safety outcomes.

Extensive efforts from transportation professionals have been made to adopt the Highway Safety Manual (HSM) [4] within the US and worldwide, recognizing its potential to enhance road safety while supporting sustainable transportation systems. The proper development of Crash Modification Factors (CMFs) not only provides practitioners with effective tools to identify and implement the most cost-efficient safety treatments but also contributes to broader sustainability objectives. By reducing crash frequencies and severities, CMFs help minimize the socio-economic costs of crashes, including medical expenses, productivity losses, and environmental impacts from crash-related delays. Various methods are available for developing CMFs, with the most widely used being simple before–after studies, before–after studies with comparison groups, before–after studies with empirical Bayes, and cross-sectional methods. Each approach offers unique strengths for tailoring interventions to specific contexts, thereby enabling safer, more sustainable, and resilient transportation networks.

Srinivasan and Bauer [5] suggested that the coefficients of the variables from Safety Performance Functions (SPFs) can be used to estimate the CMF associated with a particular treatment. This approach has been criticized earlier by Harwood et al. [6], who indicated that “regression models are very accurate tools for predicting the expected total crashes experience for a location or class of locations, but they have not proved satisfactory in isolating the effects of individual geometric or traffic control features”. Srinivasan et al. [7] concluded that in situations where before–after evaluations are not feasible, the coefficients of the variables from SPFs can be used to estimate the CMF associated with a particular treatment. The cross-sectional method requires the development of crash prediction models (i.e., SPFs) for calculation of CMFs. The models are developed using the crash data for both treated and untreated sites for the same time period (3–5 years) with a large sample size. Recently, in 2022, Carter et al. [8] released a report by the National Cooperative Highway Research Program (NCHRP), which provides broader guidelines for the development and application of CMFs.

Road lighting is designed and installed for societal and security reasons at night, so it is hard to identify the exact benefits from the implementation of this treatment on the freeway. Previous studies showed that highway lighting will improve safety on the roads by an average of 20% [9,10]. Nighttime crashes are more risky than daytime crashes, because the dark roads can reduce the visibility, maneuverability, or proper responses in dangerous locations. Zein [11] showed that streetlights reduced nighttime total crashes by 10–40%. Abuzwidah et al. [12] evaluated the effect of lighting on nighttime crashes on 4-lane rural expressways using the advanced cross-sectional method. The results proved that lighting has a positive impact on safety by approximately 30% on nighttime total crashes and a reduction of 27% on nighttime fatal and injury crashes.

Shoulder rumble strips are considered a proven and effective safety treatment, since they provide drivers tactile vibration that allows recovery and prevents road departure events. Wu et al. [13] found that shoulder rumble strips reduced crashes by 7%. Patel et al. [14] showed that shoulder rumble strips enhanced safety by 13% in total single-vehicle run-off-road (SVROR) crashes and 18% in fatal and injury SVROR crashes. Carrasco et al. [15] presented that shoulder rumble strips reduced SVROR by 20–72%.

Sustainability frameworks are increasingly recognized as essential in designing transportation systems that prioritize safety, efficiency, and environmental responsibility. The integration of renewable energy technologies, such as solar-powered lighting, aligns with global sustainability goals by reducing greenhouse gas emissions and minimizing dependence on non-renewable energy sources [16,17]. Research has demonstrated the potential of solar-powered lighting in improving road safety outcomes while addressing energy efficiency challenges [18,19]. Furthermore, sustainable infrastructure frameworks emphasize the role of energy-efficient designs in fostering long-term resilience and cost-effectiveness in transportation networks [20]. These frameworks provide a foundational context for understanding how renewable energy technologies can complement traditional road safety intervention.

Previous international studies demonstrated considerable variability in CMF values for shoulder rumble strips and roadway lighting, depending on the road type, crash type, environmental conditions, and methodological approach. For example, Torbic et al. [21] reported a CMF of 0.84 for similar treatments on two-lane rural highways in the United States. However, much lower CMF values have been reported in other contexts: Meuleners et al. [22] reported a 58% reduction in total crashes (CMF ≈ 0.42) and even larger reductions in casualty crashes (80% reduction) after implementing edge-line rumble strips in Western Australia. With respect to roadway lighting, Wanvik [23], in a comprehensive meta-analysis based on Dutch crash data from 1987 to 2006, reported a CMF of 0.50 for nighttime injury crashes on rural roads with added lighting. In contrast, Abuzwidah et al. [12] found that lighting has a positive impact on safety by approximately 30% on nighttime total crashes and a reduction of 27% on nighttime fatal and injury crashes. These differences are further influenced by ambient lighting conditions, vehicle speeds, and driver behavior, which can vary significantly between countries.

The reviewed methodologies for developing Crash Modification Factors (CMFs) highlight a variety of approaches, each tailored to specific contexts and data availability. However, the substantial variation in CMF values for interventions such as road lighting and shoulder rumble strips underscores the necessity of developing localized CMFs to ensure accurate and context-sensitive evaluations. This study aims to develop localized CMFs for these interventions in Saudi Arabia, assess their safety and economic impacts, and explore their alignment with broader sustainability objectives. The research incorporates local traffic conditions, roadway characteristics, and crash data into the development of CMFs, providing transportation professionals with accurate tools to quantify the safety benefits of these treatments. The manuscript is structured as follows: the Section 2 describes the data collection, data preparation, and analysis approaches, the Section 3 presents the safety and economic evaluation findings, and highlights the implications of these findings for resource optimization and sustainable transportation systems, and the Section 4 summarizes the main contribution and highlights the key findings.

2. Materials and Methods

The CMFs and cost-effectiveness of shoulder rumble strips and road lighting treatments on freeways were developed employing the steps demonstrated by the flowchart in Figure 1 and the procedures described underneath. The procedures contain the two major steps, including data collection and preparation, and the methodology. The methodology describes the CMF calculations and the economic analysis.

2.1. Data Collection and Preparation

The KSA is divided into 13 administrative regions, as shown in Figure 2. A significant effort was invested in the data collection and preparation process since this study covers three regions from different directions in the KSA, and these regions include the Riyadh, Makkah, and Eastern regions.

The selection of Riyadh, Makkah, and the Eastern Province as the study regions is based on their significance in representing the diversity of road conditions and traffic patterns in Saudi Arabia. Together, these regions account for approximately 67% of the Kingdom’s total population [24], making them critical areas for understanding road safety challenges and interventions. Furthermore, the availability of comprehensive and reliable data from these regions facilitated robust analysis, ensuring the development of accurate and localized CMFs. This diversity in road types, traffic conditions, and crash data enhances the generalizability of the findings to other regions within the Kingdom.

The comprehensive approach ensured the inclusion of diverse roadway and traffic conditions, enhancing the reliability and applicability of the findings. The identification of road safety interventions for analysis in this study was guided by the following criteria:

• Availability of clear information about the locations of such interventions and dates of installation.
• Availability of clear information about the type/severity of crashes related to that intervention.
• Availability of a sufficient sample size for the analysis of such interventions.
• For before–after studies, the date of installation should be during 2018, 2019, or even 2020.

In addition, SRSs and road lighting were selected because they represent the two most commonly deployed safety countermeasures on rural highways by the Ministry of Transport and Logistic Services (MoTLS). Beginning in 2018, the MoTLS began a program for implementing systemic road safety countermeasures that included SRSs. According to the analysis of data, SRSs will be evaluated using observational before–after studies with comparison groups with data related to SVROR crashes. Lighting will be evaluated using a cross-sectional method with data related to nighttime crashes.

In the before-and-after and cross-sectional evaluations, annual average daily traffic (AADT) data from MoTLS permanent counters were compiled and, where gaps existed, supplemented by targeted field counts. These traffic volumes were then linked to the site-specific road geometry and crash history for each treated segment. Untreated control sites were selected through strict matching on geometric attributes (e.g., shoulder width and lane count), AADT ranges, and pre-intervention crash frequencies to ensure that the observed safety effects could be attributed to the countermeasures rather than underlying traffic differences.

The development of CMFs using before–after studies with comparison group and cross-sectional methods requires roadway characteristics data, traffic volume data, and crash data. Those data are essential to develop CMFs. Therefore, a 5-year period of crash, traffic, and roadway characteristics data were collected for this study. The data spanned the pre-COVID-19 period (2017–2019) and post-pandemic conditions (2021–2022). The analysis considers both pre-COVID-19 and post-pandemic periods to capture potential variations in traffic patterns. During the pandemic, a nationwide curfew was imposed in Saudi Arabia, significantly reducing overall traffic volumes, particularly during evening and nighttime hours. This reduction likely led to a temporary decrease in crash frequencies and altered the distribution of crash types. To eliminate pandemic-related distortions, we have omitted all months subject to COVID-19 mobility restrictions (March–December 2020). As a result, our analysis spans January 2017 through February 2020 for the pre-pandemic period and January 2021 through December 2021 for the post-pandemic period. By excluding this interval, we ensure that our datasets accurately reflect normal conditions with respect to traffic volume and crashes, free from the distortions introduced by lockdowns and curfews. The data for this study were obtained from different sources, which are mainly the Highway Patrol, MoTLS, and National Road Safety Center (NRSC). In addition, there was a need to collect more traffic volume data and gather some roadway characteristics information in the three regions to develop reliable CMFs.

Table 1. Descriptive statistics of main variables in the three regions.

Region	Variables		Freeways
Makkah	Number of segments		316
	Total length covered [km]		679.15
	Total number of crashes		986
	AADT [veh/day]	Mean	19,925
		σ	14,751
	Segment lengths (m)	Mean	2149.2
		σ	1649.4
Riyadh	Number of segments		427
	Total length covered [km]		848.05
	Total number of crashes		2020
	AADT [veh/day]	Mean	19,386
		σ	8098
	Segment lengths (m)	Mean	1986.1
		σ	1067.2
Eastern	Number of segments		139
	Total length covered [km]		335.1
	Total number of crashes		712
	AADT [veh/day]	Mean	30,219.11
		σ	5278.31
	Segment lengths (m)	Mean	2410.76
		σ	911.71

presents descriptive statistics for the main variables in the three regions.

Regarding other variables, the study included many other variables, such as the number of lanes, median width, insider and outsider shoulder widths, driveway density, degree of curvature, curve density, proportion of roadside/median barriers, and posted speed. These variables were used to develop SPFs and CMFs using both observational before–after studies or cross-sectional analysis.

SRSs are installed to prevent SVROR crashes.

Table 2. Severity of run-off road crashes (2017–2019).

Region	Total Run-Off Road Crashes	No. of Fatal Crashes	No. of Fatalities	No. of Injury Crashes	No. of Injuries
Riyadh	650	14	21	200	292
Makkah	541	38	48	253	435
Eastern	380	12	14	79	94

shows the severity of run-off-road (ROR) crashes on freeways in the three regions of the KSA. It can be noticed that in the Makkah region, more than 7% of SVROR crashes are fatal, around half cause injuries, and 46% only cause property damage.

The crash data were collected for the before and after periods, with a focus on fatal and injury SVROR (FI SVROR) crashes.

Table 3. Treatment and comparison groups for rumble strips on freeways.

Group	Total Length (km)	Before Crashes (FI SVROR)	After Crashes (FI SVROR)
Treated	160	34	9
Comparison	168	45	23

shows a summary of the collected data for the before and after periods of the treated and comparison groups.

Regarding lighting on freeways, a large amount of data has been collected (2017–2019) in the three regions for freeways with and without lighting to be able to develop a regression model. A cross-sectional method is used since the date of light installation is unknown in some locations and before 2017 for other locations, where before data is not available.

2.2. Methodology

2.2.1. CMF Calculations

The method used to measure the impact and effectiveness of engineering interventions will enable obtaining values of the CMFs. This will be the basis for building a “National Clearinghouse of CMFs”. Therefore, the selection and application of the appropriate models that enable obtaining the CMFs are very important. The HSM 2010 [4] suggests several methods for safety effectiveness evaluation and obtaining CMFs. After a comprehensive review, the use of the following two methods was proposed for this study based on available data:

• Before–after studies with comparison groups;
• Cross-sectional studies.

The use of the first method, which is a before–after study with comparison groups, has many advantages for its statistical ease. However, according to the HSM 2010 [4], it requires considerable experience to determine the treatment group and the comparison group. One key concern in before–after studies with comparison groups is the regression-to-the-mean (RTM) effect, which can occur when sites are selected based on unusually high crash rates in the “before” period. To mitigate this, the analysis incorporates multiple years of crash data and compares treated sites with similar untreated sites in terms of traffic volume, roadway characteristics, and crash history. This approach reduces the influence of RTM on the results. The variability in crash data due to external factors, such as seasonal changes, variations in traffic volume, or environmental conditions, is another limitation. This study addresses these issues by using a sufficiently large sample size and including diverse sites across different regions, ensuring that the findings are representative. Additionally, statistical modeling accounts for these variations, enhancing the robustness of the results. It also requires that before and after data are available for both treated sites and untreated comparison sites. To ensure comparability between treated and comparison sites during the “before” period, we applied a three-step matching procedure. First, comparison segments were required to have an AADT within ±10% of the corresponding treated site. Second, roadway characteristics were aligned by matching the number of lanes, median type and width, and shoulder presence and width. Finally, comparison sites were selected to exhibit crash-frequency and severity distributions within ±10% of those observed at the treated segment. This rule-based approach guarantees that the exposure, geometry, and safety history are closely balanced, thereby isolating the true effect of the interventions. When treatment installation dates are not available or when crash and traffic volume data for the period prior to treatment implementation are not available, then it is recommended by the HSM to apply cross-sectional studies.

According to the HSM 2010 [4], there is no step-by-step methodology for the cross-sectional safety evaluation method because this method requires model development rather than a sequence of computations. In implementing the cross-sectional safety evaluation method, all of the crash, traffic volume, and site characteristics data (including data for both the treatment and nontreatment sites) are analyzed in a single model.

The procedure to be followed for estimating Crash Modification Factors using observational before–after studies with comparison groups will be that suggested by Gross et al. [25] based on the work of Hauer [26].

In that procedure, CMF is estimated as follows:

$$\mathrm{CMF} = (N_{observed, T, A}/N_{expected, T, A})/(1+(\mathrm{Var} \left(N_{expected, T, A}\right)/N_{expected, T, A}2\left)\right)$$

(1)

where

• $N_{observed, T, B}$ = the observed number of crashes in the before period for the treatment group.
• $N_{observed, T, A}$ = the observed number of crashes in the after period for the treatment group.
• $N_{observed, C, B}$ = the observed number of crashes in the before period in the comparison group.
• $N_{observed, C, A}$ = the observed number of crashes in the after period in the comparison group.

And

$$N_{expected, T, A}=N_{observed, T, B}(N_{observed, C, A}/N_{observed, C,B})$$

(2)

$$\mathrm{Var}\left(N_{expected, T, A}\right)=N_{expected, T, A}2(1/N_{observed, T, B}+1/N_{observed, C, B} +1/N_{observed,C, A})$$

(3)

The variance of CMF would be as follows:

$$\mathrm{Variance}(\mathrm{CMF}) = {\displaystyle \frac{{CMF}^2[\left(\frac{1}{N_{observed, T, A }}\right) + (\frac{Var\left(N_{expected, T, A}\right)}{N_{expected, T, A}2}\left)\right]}{{[1 + Var(N_{expected, T, A})/N_{expected, T, A}2]}^2}}$$

(4)

Lastly, to construct a confidence interval, the standard error is scaled by the appropriate quantile (95%) and then added to and subtracted from the CMF estimate. This calculation follows the formula given in Equation (5).

$$95\% \mathrm{Confidence} \mathrm{Interval}=\mathrm{CMF} \pm 1.96\times Standard Error$$

(5)

2.2.2. Economic Analysis

The economic analysis of safety interventions aims to evaluate the effectiveness of countermeasures. This involves converting the expected change in average crash frequency into monetary values and comparing it to the cost of implementing the countermeasure. The process results in two key metrics: the benefit–cost ratio (BCR) and the Net Present Value (NPV). The NPV represents the difference between the present value of benefits and the present value of costs associated with a safety improvement project [27].

The present value (PV) is an economic concept that consolidates costs or benefits occurring at different time periods into a single value at the present time. This is achieved by applying an appropriate discount factor to each cost or benefit, which converts them into present values. The discount rate, expressed as an annual percentage, reflects the rate of return that money could earn over the analysis period. This analysis period is commonly referred to as the service life of the countermeasure. The service life, determined by local conditions, is used to calculate the present value of the proposed countermeasure’s benefits and costs [28].

A reasonable discount rate of 5% can be used for safety improvements’ economic appraisal considering the current values of interest rates in 2023.

The following is the equation to calculate the PV:

$$PV=A\times {\displaystyle \frac{({1 + i)}^y-1}{i \times ({1 + i)}^y}}$$

(6)

where

• PV = present value;
• A = annual benefit (i.e., monetary value of crashes prevented) or cost;
• i = discount rate;
• y = service life of the countermeasure.

3. Results

3.1. Shoulder Rumble Strips

The results of the analysis using a before–after analysis with a comparison group show that shoulder rumble strips reduce fatal and injury SVROR crashes by 52.7% (CMF = 0.473) with a 95% CI of 0.325–0.621. When compared with factors given in the USA for longitudinal rumble strips on rural freeways, a 17% reduction in SVROR FI crashes was reported [21]. Additionally, there is considerable difference between the results obtained in the KSA and other countries (Norway, Australia, and Sweden), as reported in Vadeby and Anund [29]. It should be noted that there is a considerable variation in the values provided in the CMF clearinghouse, which shows the importance of developing local CMFs.

3.2. Presence of Lighting

With respect to the presence of lighting, a crash prediction model (i.e., SPF) was developed using nighttime crashes only to calculate the CMF. Information on sections provided with lighting was gathered from data on homogeneous segments. The model was developed using the crash data for both treated and untreated sites for the same time period. Regression analysis of nighttime crashes on freeways in the Makkah, Riyadh, and Eastern regions was conducted to obtain a CMF for lighting intervention for all regions combined. The model obtained is as follows:

$$\begin{array}{l}\mathrm{Average} \mathrm{crash} \mathrm{frequency} = \mathrm{n} \times \mathrm{L}\times \mathrm{EXP}\left(2.497\right)\times \mathrm{EXP}\{-0.273\times (\mathrm{Lighting}\left)\right\} \times \mathrm{EXP}\{0.014\times (\% \mathrm{cars}\\ \mathrm{over} 140 \mathrm{km}/\mathrm{h}\left)\right\}\times \mathrm{EXP}\{-0.301 \times (\mathrm{Median} \mathrm{Barrier}\left)\right\}\times \mathrm{EXP}\{-0.124 \times (\mathrm{Median} \mathrm{width}\left)\right\}\times \mathrm{EXP}\{-0.584 \times \\ (\mathrm{Inner} \mathrm{shoulder} \mathrm{width})\}\end{array}$$

All variables have negative coefficients except for the percent of car driving over 140 km/h, which has a positive coefficient, indicating that the predicted crash frequencies would increase with an increase in this percentage. The percent of cars driving over 140 km/h is a measure of exposure to traffic. An increase in traffic exposure is expected to increase the crash frequency. The presence of lighting is a dummy variable with a value of zero or one. The CMF can be estimated as the ratio of the predicted number of crashes when there is lighting to the predicted number of crashes when there is no lighting:

$$\mathrm{Lighting} \mathrm{CMF} = {\displaystyle \frac{EXP\{-0.2727 \times (1.0\left)\right\}}{EXP\{-0.2727 \times (0.0\left)\right\}}}=\mathrm{EXP}\{-0.2727\times (1.0–0.0)\}$$

Lighting CMF = EXP(−0.272757) = 0.76

It is clear that the treatment has a positive effect on nighttime crashes on freeways, since it is expected to reduce nighttime crashes by 24%. This result is comparable with many previous studies [9,10].

3.3. Economic Impacts of the Interventions

Regarding the economic feasibility of the SRS intervention on freeways in Saudi Arabia, discussions with the MoTLS determined that a service life of 7 years is appropriate. This estimate aligns with the Federal Highway Administration (FHWA)’s recommended service life for rumble strips, as they are typically re-installed during overlays or major maintenance. The annual reduction in fatal and injury (FI) crash costs due to SRS implementation was aggregated to calculate the total cost savings over the pavement’s lifespan.

Based on input from MoTLS maintenance contractors, the installation cost of rumble strips is SR4000 per kilometer per strip. For freeways, which require four strips, the total installation cost is SR16,000 per kilometer. The average costs associated with fatal and injury crashes, as reported by the National Road Safety Center [30], were used in the analysis to quantify the economic benefits of crash reductions, and the values are as follows:

• Fatal crash = SR5,984,338;
• Serious injury crash = SR1,408,011;
• Minor injury crash = SR91,580.

It should be noted that around 80% of the injuries are severe injuries and 20% are minor injuries. Therefore, the cost of an injury crash will be as follows:

Cost of Injury crashes = 0.8 × 1,408,011 + 0.2 × 91,580 = SR1,144,725.

Accordingly, the monetary annual benefit of the expected fatal crash reductions (CMF = 0.473) is as follows:

Average Fatal ROR crashes per year per km × reduction in crashes × cost of fatal crash
= 0.0044 × 0.527 × 5,984,338 = SR13,876.5

The monetary annual benefit of the expected injury crash reductions is as follows:

Average Injury ROR crashes per year per km × reduction in crashes × cost of injury crash
=0.0417 × 0.527 × 1,144,725 = SR25,156

Thus, the total monetary annual benefit from F + I crash reductions is as follows:

13,876 + 25,156 = SR39,032/km/yr.

Thus

$$\mathrm{PV} \mathrm{of} \mathrm{benefits}=39,032 \times {\displaystyle \frac{{(1+0.05)}^7-1}{0.05\times {(1+0.05)}^7}}=\mathrm{39,032} \times 5.789=\mathrm{SR}\mathrm{225,958}/\mathrm{km}$$

Present value of costs of rumble strips = SR16,000/km

Therefore,

BCR = 225,958/16,000 = 14.12.

The analysis reveals that the benefit–cost ratio for shoulder rumble strips on freeways is 14.12, indicating a highly favorable outcome where the benefits significantly outweigh the costs. This substantial ratio underscores the economic efficiency and effectiveness of SRSs as a safety intervention, making them a valuable investment for enhancing road safety and supporting sustainable transportation infrastructure.

With respect to lighting, on average, there will be 20 double poles per kilometer (a pole every 50 m in each direction). Therefore, the cost of delivery and installation is as follows:

20 × 2 × 14,000 = SR560,000/km.

The annual cost of power supply and maintenance per kilometer is (2 × 200 × 20 = SAR 8000 per year per kilometer) with a service life of 15 years, which is conservative since some studies use 20 or more years of service life.

Considering Road No. 40 between km 500 and km 600 (with no lighting), where the nighttime crashes between 2017 and 2019 were six fatal crashes and 53 injury crashes. Therefore, the average nighttime fatal and injury crashes per km per year are 0.02 and 0.176, respectively.

The monetary benefit of the expected fatal crash reductions (CMF = 0.76) is as follows:

0.02 × 0.24 × 5,984,338 = SR29,264

The monetary benefit of the expected injury crash reductions (CMF = 0.76) is as follows:

0.176 × 0.24 × 1,144,725 = SR48,353

Thus, the total monetary benefit from F + I crash reductions = 29,264 + 48,353 = SAR 77,617/km/yr.

As mentioned before, as the service life of lighting poles is 15 years and the discount rate is 5%, then,

$$\mathrm{PV} \mathrm{of} \mathrm{benefits}=77,617 \times {\displaystyle \frac{{(1+0.05)}^{15}-1}{0.05\times {(1+0.05)}^{15}}}=\mathrm{77,617} \times 10.38 = \mathrm{SR}\mathrm{805,666}$$

Present value of power consumption and maintenance = 8000 × 10.38 = SR83,040/km

Present value of costs = 560,000 + 83,040 = SR643,040/km

Thus,

BCR = 805,666/643,040 = 1.253

The benefit–cost ratio for lighting on freeways is calculated at 1.253, demonstrating that the benefits outweigh the costs, making the intervention economically viable. This value remains above one even when a conservative service life of 15 years is applied, highlighting the robustness of the analysis. Moreover, adopting solar-powered lighting systems could significantly enhance the benefit–cost ratio, particularly in Saudi Arabia, where abundant sunshine provides an ideal opportunity for integrating renewable energy solutions into sustainable transportation infrastructure.

In summary, the results show clear, actionable insights for road safety policymakers in Saudi Arabia. The significant reduction in fatal and injury SVROR crashes due to the use of shoulder rumble strips, with a CMF of 0.473 and a BCR of 14.12, demonstrates a highly effective and economical treatment. This suggests that SRSs could be prioritized in national safety programs, particularly on high-speed rural freeways.

Likewise, road lighting treatments—while showing a more modest crash reduction of 24% (CMF = 0.760)—are particularly relevant for improving nighttime driving conditions. The adoption of solar-powered lighting systems not only enhances safety but also supports broader environmental and energy efficiency goals, aligning with the KSA’s Vision 2030 and global sustainability targets.

By integrating both safety and sustainability considerations, these findings support the development of comprehensive road safety strategies that are not only technically valuable but also cost-effective and environmentally responsible. This can help optimize limited infrastructure investment resources while maximizing public safety benefits.

These practical implications emphasize that even simple strategic engineering interventions, compared to other major treatments, could lead to significant improvements in safety, sustainability, and economic performance.

3.4. Limitations

Although the study revealed several key findings, there are still some limitations to this study. Firstly, the essential data were collected and aligned with supplemental data but resulted in a limited number of factors; further variables can be considered to capture their impacts. Secondly, as the SPF was developed using nighttime crashes only to calculate the CMF of the presence of lighting, exploring regional variations in crash severity could provide deeper insights into the contextual factors affecting safety outcomes. In terms of the methodology, this study adapted the negative binomial to develop the SPF as in the HSM, and the before–after studies with comparison groups for the CMF development, but it may be beneficial to conduct these using more advanced approaches and in-depth analysis of the limitations related to the data and methods. Lastly, this study considered only the freeway; future studies could consider other road classifications. Other road classifications such as multilane or two-way two-lane roads were not included. It is essential to note that road classification could significantly impact the safety performance of countermeasures such as shoulder rumble strips and road lighting. For instance, the Highway Safety Manual provides separate SPFs for different roadway types, reflecting their distinct traffic characteristics, design standards, and crash patterns. Consequently, the findings of this study may not be directly transferable to non-freeway contexts. Future research should expand the dataset to include various road types and develop context-specific SPFs to better understand the broader applicability of these safety interventions.

4. Conclusions

This study represents a pioneering effort at the national level in Saudi Arabia, implementing Part D of the HSM to enhance the methodology for selecting road safety countermeasures on freeways. By developing localized CMFs based on region-specific data, this research provides valuable tools for transportation professionals to make informed, data-driven decisions.

The study estimated CMFs for two key safety interventions on freeways: SRSs and road lighting. The evaluation of SRSs, conducted using a before–after analysis with a comparison group, focused on fatal and injury SVROR crashes. Lighting, evaluated through a cross-sectional study due to data limitations on installation dates, targeted nighttime crash reductions. The data, spanning the pre-COVID-19 period (2017–2019) and post-pandemic conditions (2021–2022), provided a robust foundation for analysis.

The results showed that SRSs reduce fatal and injury SVROR crashes by 52.7% (CMF = 0.473, 95% CI: 0.325–0.621) and are highly cost-effective, with a benefit–cost ratio of 14.12. Road lighting reduces nighttime crashes by 24% (CMF = 0.760) and demonstrates a marginal benefit–cost ratio of 1.24. However, adopting solar-powered lighting can significantly enhance the environmental and economic sustainability of this intervention, especially given Saudi Arabia’s abundant sunshine.

While the general trends observed in this study align with international findings, the contribution lies in developing localized CMFs and economic evaluations tailored to Saudi Arabian highways. These results support the efforts in the KSA towards enhancing traffic safety by offering regionally calibrated data for evidence-based decision-making. This research also emphasizes the alignment of engineering solutions with sustainability principles. By reducing crash-related socio-economic costs and integrating renewable energy technologies, the interventions contribute to safer, more equitable, and environmentally responsible transportation systems. This study underscores the importance of a multidisciplinary approach in advancing road safety and sustainable mobility on both the national and global scales.