Speeds on Rural Interstate Highways Relative to Posting the 40 mph Minimum Speed Limit
The relevance of posting the 40 mile per hour (mph) minimum speed limit on the Interstate Highway System has been increasingly called into question since the National Highway System Designation Act of 1995 repealed the federally sanctioned maximum speed limit. In this study, data were collected on major interstate highways in Florida to evaluate speed distribution relative to the 40 mph posted minimum speed limit. The data revealed that the 15th percentile speed at all sites was 60 mph or above on both four-lane and six-lane highway sections. The analysis showed that the average speed at all sites was approximately 5 standard deviations above the 40 mph minimum. The coefficient of variation ranged from 7% to 11%, while the trimmed variance analysis showed that vehicles traveling below 55 mph contributed insignificantly to the variation in traffic speeds. A comparison of data collected before the speed limit rose from 65 mph to 70 mph showed that the average speed increased by 5 mph, while the variances did not change significantly. The coefficients of variation, however, increased significantly. The results reported here suggest that speed variability at the lower end of the distribution is not a significant factor in traffic operating characteristics on Florida rural interstate highways.
KEYWORDS: Speed limit, speed variation, highways.
The decades-old practice of posting minimum speed limits on rural interstates and other limited access highways is predicated on the desire to reduce vehicle conflicts caused by speed variability in a traffic stream. The relevance of the 40 mile per hour (mph) posted minimum speed limit found on the Interstate Highway System is increasingly being called into question in light of the National Highway System Designation Act of 1995, which repealed the federally sanctioned maximum speed limit of 65 mph on rural highways. Most states, including Florida, then raised the maximum speed limits, and by the end of 1997, most parts of Interstates 4, 10, 75, and 95 in Florida posted 70 mph, which is the maximum speed allowed by the Florida state statutes. While the maximum speed limit fluctuated over time, the minimum did not and, in Florida, the 40 mph limit was in effect and posted across many sections of rural interstate highways, even when the U.S. Congress required states to lower the speed limit to 55 mph in 1974.
With such a wide (30 mph) gap between maximum and minimum speed limits, it is logical to question the relevance of the 40 mph posted minimum. If the review of the current speed distribution shows that the 15th percentile speed is much higher than the 40 mph posted minimum, perhaps the minimum speed needs to be increased or rescinded. Also, it is important to know if the continued posting of the 40 mph minimum speed limit results in the increase in speed variability on rural interstate highways. A review of traffic operations on sections of Florida highways may provide answers to these questions.
The promise behind the posting of minimum speed limits on interstate highways was to reduce interactions between fast and slow moving vehicles. Many states based their minimum speed limits on the Uniform Vehicle Code (UVC) published by the National Committee of Uniform Traffic Laws and Ordinances (National Committee 1954). The UVC stipulated that minimum speed limits be established on highways whenever traffic and engineering investigations concluded that slow-moving vehicles consistently impeded the normal flow of traffic on the highways.
Studies showed that, by 1962, many states had adopted slow speed laws in their statutes in compliance with the UVC (National Committee 1964). Florida was among the states adopting slow-speed provision, making 40 mph the minimum on the four-lane interstate system, the Turnpike, and defense highways. Basically, the Florida statutes made it illegal to drive at a slow speed that impedes the normal and reasonable flow of traffic on rural highways.
The literature reveals that, in the early 1960s, 41 states and the District of Columbia instituted slow-speed laws in verbatim or significantly conforming with the UVC, while the remaining 9 states did not add minimum speed regulations in their codes. Like Florida, Georgia and South Dakota statutes explicitly stated that the minimum speed limit was 40 mph, while Michigan and North Carolina maintained a 45 mph minimum speed rule on their interstate highways.
A 2003 survey of minimum speed practices in different states conducted for the Florida Department of Transportation showed that, following the 1995 National Highway System Designation Act, 43 states raised the maximum speed limit on their Interstate Highway System roads (Mussa 2003). However, the posted minimum speed on these systems did not change. In fact, the survey showed that 14 states still use 40 mph minimum speed limit signs, 10 states use 45 mph, and 1 state uses 55 mph. Furthermore, the survey showed that 25 states do not post minimum speed limit signs. Some respondents in states that do not post minimum speed limit signs indicated that slow driving is not a big problem on their highways and if a need arose for enforcement, various rules in their state statutes, such as "impeding traffic flow," can be used to warn or cite slow drivers.
UNDESIRABLE EFFECTS OF SPEED VARIABILITY
Posting a minimum speed limit was and still is motivated by the desire to reduce speed variability in a traffic stream and its attendant consequences in efficiency and safety of traffic operations. Numerous studies have documented the negative effects of speed variability.
In determining the extent to which the 55 mph federally sanctioned maximum speed limit affected safety, a Transportation Research Board (TRB) study found that the probability of crashes occurring increases as the speed variance rises. The study showed that speed variation causes significant lane changing and passing maneuvers, which are known to be potential sources of conflicts and crashes (TRB 1984). The significance of speed variance was observed by developing a fatality model that included highway safety characteristics such as traffic density, percentage of vehicles exceeding 65 mph, percentage of teenagers, and enforcement activity, as well as speed variance and average speeds. The TRB model revealed that speed variance had a statistically significant effect on fatality rates-states with wider variances in vehicle speed on the highway tended to have higher fatality rates. The study further found that the mean speed only affected the severity of crashes. Holding the effect of speed variance constant in the model presented no statistically significant relationship between the fatality rate and any other speed variables. The study concluded that controlling speed variance could be an effective tool in improving highway safety.
Another study of 36 crashes that occurred on Indiana highway 37 indicated that the crash involvement rates per million vehicle-miles of travel were higher for vehicles whose speeds were below and above the mean speed (West and Dunn 1971). After removing data on all crashes related to turning maneuvers, the authors found that the crash risk associated with vehicles traveling faster or slower was more than six times the involvement rates at the mean speed. The West and Dunn findings were supported by Hauer (1971) who developed a mathematical model to correlate accident involvement rates and vehicle travel speeds. Hauer found that the imposition of a minimum speed limit on highways was two to three times as effective as an equivalent maximum speed limit in reducing the frequency of overtaking and thereby crash involvement rates. Hauer suggested that the relationship between vehicle speed deviations and crashes might be due to a higher incidence of passing maneuvers from which the vehicle passes or is passed by another vehicle-a situation caused by the presence of slower vehicles impeding fast vehicles in the traffic stream.
Lave (1985) found that the major highway safety benefits obtained after the enactment of the 1974 National Maximum Speed Limit Act-which reduced the maximum speed limit on interstate highways to 55 mph-were due to the reduction of speed variance rather than average speed. The author argued that a reduction in speed variance was realized because speed differences between slow and fast moving vehicles were reduced enough to cause a uniform flow of traffic on interstate highways. Thus, with small speed variances there are fewer passing and overtaking maneuvers, eventually leading to the reduction in the potential for conflicts and crashes. Lave concluded that slow drivers are just as dangerous as fast drivers and thus posting minimum speed limits is desirable so as to reduce speed variance in a traffic stream.
RESEARCH AGENDA FOR A MINIMUM SPEED LIMIT
The posting of higher maximum speed limits on rural interstate highways necessitates an evaluation of the relevance of posted minimum speed limit signs that existed prior to raising the maximum speed. Some studies (e.g., West and Dunn 1971; Hauer 1971; and Lave 1985) documented that posting the minimum speed limit has the beneficial effect of smoothening traffic flow by removing perturbations caused by speed differences.
While evidence obtained from past research shows that vehicle speed variability contributes to crashes, it is a big and unsubstantiated leap to say that posting 40 mph minimum speed limit signs on a highway with a 70 mph maximum speed limit, as is the case on the Florida rural Interstate Highway System, contributes to large differences in vehicle speeds. The effect of the 40(min)/70(max) seeming mismatch can be evaluated through a carefully designed field study in which driver characteristics and the resulting operating speeds are observed over a long period of time on highway sections with similar geometrics and traffic characteristics but with some having the 40 mph minimum posted and others not having the minimum posted. Furthermore, knowing whether the minimum speed limit should be increased above 40 mph and by how much, given that the maximum speed limit has been raised from 65 mph to 70 mph, would also be useful. To obtain this information, a study would require experimental highway sections with the desired minimum speed limit signs posted.
This study aimed at evaluating operating speed characteristics on the Florida Interstate Highway System where 40 mph minimum speed limit signs are posted. It would have been desirable to conduct a study designed as described above but control sites with no minimum speed limit signs were not available. An experimental site with 40 mph minimum speed signs removed or covered can be created for conducting a longitudinal study where both operational and traffic crash data are collected and later compared with the current conditions. However, creating such sites has legal implications that are difficult to resolve at this time. Thus, this study was limited to the following: determining how speed characteristics deviate from the 40 mph limit, and determining the speed variability that resulted before and after the limit was raised.
Note that the relevance of the 40 mph minimum speed limit is analyzed in this paper from the operational standpoint only. Certainly, law enforcement personnel would prefer to have these signs erected to provide support for warning or citing slow moving drivers; the "impeding traffic flow" criterion may be less useful for enforcement purposes.
There are four interstate highways in Florida, Interstates 4 and 10, oriented in the east-west direction, and Interstates 75 and 95, which go in a north-south direction. In addition, the Florida Turnpike is a tollway from central to south Florida oriented in the north-south direction.
Site selection targeted rural sections of these roads where minimum speed limit signs are posted. The established site selection criteria required choosing sites where the geometric characteristics produced the highest free-flow speed possible, that is, sites devoid of horizontal and vertical curves, sustained grades, or other geometric constraints. Another criterion used was to select sections with telemetered traffic monitoring stations that collect traffic flow data on volume, occupancy, and individual vehicle speeds on a 24-hour basis throughout the year. The Florida Department of Transportation operates and maintains these sites. We could not select a site on Interstate 4 because the Tampa-Orlando-Daytona Beach corridor, through which this highway runs, is heavily congested throughout the week with few periods in which free-flow speeds are attainable. Table 1 shows the study sites selected based on the established criteria discussed above.
As part of the data collection strategy, the project team drove through the entire Interstate Highway System to observe geometrics and traffic operating conditions. In addition, the project team evaluated over 320 telemetered traffic monitoring sites to determine their locational suitability in relation to the research objective of evaluating speed characteristics. The field review resulted in choosing sites described in table 1. The elements of the data-collection plan including the data quality checks are explained below.
Individual Vehicle Records
Telemetered traffic monitoring stations use loop detectors that provide individual vehicle records composed of the exact time of passage of a vehicle, its speed, the lane of passage, the number of axles and axle spacing, vehicle length, and, in some stations, an individual vehicle's axle weight. A cursory review of the speed characteristics at most sites indicated that there were minor differences between weekend and weekday traffic speed distribution. Thus, data from all sites were collected on weekdays in good weather conditions and dry pavement. The integrity of the data was verified by checking for errors.
Data Error Checks
Logic checks on the recorded data elements were applied to the raw data files downloaded from the count stations. Typical data errors included improper recording of speeds and loop failures on some lanes. We set an initial criterion that if more than 5% of the data were bad, the dataset for the whole day was discarded and another day's data were downloaded. The accuracy of individual vehicle speeds was checked by relating vehicle length and its corresponding recorded speed. When the length of the vehicle was missing, it was assumed that the vehicle did not cross both loops in the speed trap thus suggesting that the recorded individual vehicle speed was erroneous. The number of records with missing speed or length was used to check the percentage of usable counts with respect to raw data elements and finally to decide whether the data for that particular day was within acceptable limits needed for further analysis.
Next, outliers were removed prior to performing the statistical analyses. All outliers, defined as data points that were inconsistent with the general trend of the data elements, were eliminated by a computer program developed for this purpose. The computer program discarded data points showing zero speed or speed greater than 120 mph (the maximum speed value the equipment can record). The time slots in which data were discarded were coded as missing data. In all datasets used for further analysis, the percentages of data coded as missing were less than one. All individual vehicle records were then summarized per hour and per lane and the required volume, speed, and headway statistics were calculated. Analyses proceeded only after assuring the data quality through these error checks.
Under low to moderate traffic congestion, as demand on travel lanes increases so does the need of fast moving vehicles to pass slow moving vehicles. The combination of passive and active passing maneuvers creates the potential for conflicts in the traffic stream. Higher operating speeds are generally attainable at level of service (LOS) A1 and continually decrease as the speed-volume relationship moves toward congested flow conditions. An hour-by-hour volume analysis of the 24-hour dataset was conducted at all eight sites to determine the volume distribution across the travel lanes, the percentage of trucks on each lane, and the minimum and maximum volumes and their hour of occurrence. The traffic volumes were expressed on a per-lane basis, because, in general, volume varies by lane. The average annual daily traffic, which is the gross indicator of traffic activity, usage, and need, was estimated by multiplying the 24-hour recorded volume with the adjustment factors developed by the Florida Department of Transportation Statistics Office. Table 2 shows the results of the volume analysis.
Table 2 presents the results categorized by the number of lanes on the highway (i.e., four or six lanes) and by direction of travel. Examination of the hourly variation at each site showed that the demand volumes were at their lowest from midnight to dawn hours, while the peak-hour demand occurred in the afternoon, typically from 3 p.m. to 5 p.m. with a few exceptions. The lane distribution analysis for the six-lane highway sections showed that flow rates in the middle lane were typically higher than on shoulder and median lanes. On four-lane sections, the flow rates on the shoulder lanes were higher than on the median lanes.
We also analyzed the distribution of trucks in each lane. Vehicles traveling at the low end of the speed distribution tended to be trucks, recreational vehicles, and vehicles towing trailers. Table 2 shows that truck percentages are higher on the shoulder lanes in both four-lane and six-lane sections. Note that on Sites 320 and 9904 on Interstate 75 in north Florida trucks are not allowed to travel on the median lane of three-lane (in one-direction) sections (i.e., they can only use the two outermost lanes).
A comparison of the peak-hour and 24-hour truck percentages suggests that more trucks travel during the offpeak hours. The need to change lanes and to pass some slow moving vehicles-typically trucks and RVs-is high during offpeak hours. The LOS in most of the sections was B or better during these time, thus operating speeds tend to be high due to fewer traffic interactions. With trucks and RVs typically among the slower moving vehicles, changing lanes and passing, resulting from the speed variances, might be a concern.
The analysis of speed is presented in two parts. The first part of the analysis details the central tendency of the speed data while the second part looks at the speed variability in the traffic stream. The analysis of both measures of center and dispersion takes into account the demand volume, lane of travel, and the type of vehicles-passenger cars or trucks-in the traffic stream.
Central Tendency Analysis
Figure 1 shows the 24-hour mean speed of all vehicles categorized by facility type (i.e., four-lane or six-lane highway). Examination of the graphs in figure 1 reveals that average speeds of vehicles vary from shoulder to median lanes with median lanes experiencing higher average speeds. At four-lane sites, the average speeds ranged from 66 mph to 74 mph in shoulder lanes and 67 mph to 85 mph in median lanes. At six-lane sites, the average speeds of the vehicles on the shoulder, middle, and median lanes ranged from 67 mph to 70 mph, 72 mph to 75 mph, and 75 mph to 81 mph, respectively.
Pairwise comparisons of the average speeds using a t-test showed that, on four-lane sections, average speeds differed significantly between shoulder and median lanes (p = 0.0002). Further analysis showed that the average speeds were significantly different between shoulder-middle lanes and middle-median lanes on six-lane sections (p ≤ 0.0001 and p ≤ 0.0001). These results confirm that slow-moving vehicles generally use the shoulder lanes while fast moving vehicles use the median lanes. At the prevailing LOS, it seems that the influence of traffic intensity was not a significant factor, because at six-lane sites the middle lanes carried higher volumes than shoulder lanes yet they had higher average speeds. To further understand the profile of speeds at these highway sections, table 3 presents the overall 24-hour mean speeds by lane and vehicle type. Table 3 also shows the harmonic mean speeds weighted by lane volumes and by vehicle type. The harmonic mean speeds were calculated as follows:
= the harmonic mean speed weighted by the 24-hour lane volume in lane i,
= the harmonic mean speed weighted by 24-hour vehicle type j volume,
= the 24-hour mean speed of all vehicles in lane i,
= the total 24-hour volume in lane i,
= the 24-hour mean speed of all vehicles of type j, and
= the total 24-hour volume of vehicle type j.
Table 3 also includes the straightforward average speeds of all vehicles and the trimmed mean speeds. The trimmed mean speeds were calculated by discarding the lowest 15% and the highest 15% of vehicle speeds. We statistically analyzed the significance of the difference between the speed types displayed in this table. Pairwise comparisons of and showed no and slightly significant differences (p = 0.7 and p = 0.08) between lane-based and vehicle type-based mean speeds on both six-lane and four-lane highway sections, respectively. Statistical comparisons between trimmed mean speed and average speeds in each lane indicated lack of a discernible difference in both six- and four-lane sections (p = 0.57 and p = 0.40). The non-existence of the difference between trimmed speed and mean speed shows that the presence of fast and slow moving vehicles in the speed distribution has no significant effect on the average speeds on these facilities. The average speed of the bottom 15th percentile of the vehicles was 62 mph on both facility types, while in the upper 15th percentile, the average speed of vehicles was 81 mph and 83 mph on six-lane and four-lane sections, respectively.
Speed Dispersion Analysis
The dispersion of speeds was analyzed by lane and vehicle type using the standard deviation, coefficient of variation, and 10-mph pace, which is the 10 mph speed range with the highest number of observations of vehicles in the speed distribution. In addition, as is the case in most traffic engineering design and operational analyses, the 85th and 15th percentile speeds were also calculated. The results follow.
Facility Type Speed Distribution
We computed the standard deviations of vehicle speeds and the corresponding coefficient of variation. The results showed that their values varied depending on facility type. On six-lane sections, the standard deviation of speeds ranged between 4 mph and 6 mph, while on four-lane sections the standard deviations were as high as 10 mph. Specifically, Sites 351 and 9919 showed high values of standard deviations-9 mph and 10 mph on the median lanes, respectively. The field review revealed that these two sites are on highway stretches that are longitudinally straight for at least 10 miles.
The coefficient of variation, which measures relative dispersions of vehicle speeds from the average speed, was also calculated by lane for each site. This statistic was necessary to compare speed variations by examining the magnitudes of deviation relative to the magnitude of the mean given that there were different mean speeds grouped by lane. The analysis of the coefficients of variation in each lane showed that they ranged from 5% to 14%. When coefficients of variation for adjacent lanes on each site were compared, the results showed that the differences were less than 2%. These results suggest that the scatters of the vehicle speeds from the average speed are small. Therefore, the traffic speeds are very closely clustered about the mean speeds in all sections analyzed.
Speed Distribution by Vehicle Type
On average, the results of the speed distribution analysis by vehicle type showed that passenger car speeds were higher than truck speeds by at least 1 mph on both six-lane and four-lane sections. The results further showed that the coefficients of variation did not differ significantly between passenger cars and truck speeds for six-lane highway sections but were significant on four-lane sections (p = 0.027). Figure 2 displays the results of the speed distribution analysis at the lower end of distribution.
With respect to the vehicles traveling at the lower end of speed distribution (i.e., less than 60 mph), we found that more trucks on four-lane sections traveled below 60 mph than passenger vehicles at Sites 9901, 9919, and 9928, while more passenger cars traveled below 60 mph at Sites 351 and 9932. The results were also mixed on six-lane highway sections. At Sites 320 and 9904, which are on the same stretch and approximately 70 miles apart, different patterns of vehicles traveling below 60 mph were observed. While at Site 9904 more passenger cars traveled at speeds below 60 mph, at Site 320 more trucks traveled below 60 mph. At Site 9905, more passenger cars than trucks had speeds below 60 mph in all lanes.
The results further showed that on both four-lane and six-lane sections the percentage of vehicles at each site traveling below 40 mph (the posted minimum speed limit) was approximately zero. In fact, the results showed that at all sites only 1% of the vehicles traveled below 55 mph. Both passenger cars and trucks averaged speeds below 60 mph but above 55 mph on six-lane sections. On four-lane sections, the speed of vehicles traveling below 60 mph averaged above 54 mph.
Percentile and Pace Characteristics
Table 4 displays the 15th and 85th percentile speeds in each lane, 10 mph pace speeds, and the percentages of vehicles within the pace. Analysis of percentile speeds showed that, in four-lane and six-lane sections, the 85th percentile speeds ranged from 71 mph to 94 mph and 73 mph to 86 mph, respectively, while the 15th percentile speeds ranged from 60 mph to 77 mph and 62 mph to 76 mph, respectively, depending on the lane of travel (i.e., median lanes had higher percentile speeds than shoulder lanes). Of significant interest was the 15th to 85th percentile range, because it represents the proportion of vehicles traveling close to the mean speed. At the six-lane sites, the percentile speeds ranged from 7 mph to 10 mph, 8 mph to 10 mph, and 10 mph to 12 mph on the median, middle, and shoulder lanes, respectively. The ranges for four-lane sites were 7 mph to 11 mph and 11 mph to 12 mph on the median and shoulder lanes, respectively.
Note that the results from Sites 351 and 9919 do not particularly follow the trend of other sites because of the somewhat large differences between percentile speeds at the two sites-14 mph and 19 mph, respectively. These differences could result from the straightness of the segments as well as a low volume of traffic that induces high-speed travel by some drivers. Furthermore, these two sites also showed the highest values of standard deviations. Table 4 further details that the paces ranged from the mid-60s to the mid-80s on both facility types with shoulder lanes experiencing lower pace speeds. The results in table 4 show that there is no direct relationship between the number of lanes on a highway and pace speeds.
Trimmed Variance Analysis
A trimmed variance analysis determined the contribution of slow- and fast-moving vehicles on overall speed variation. Using five different scenarios, vehicles traveling slower than 40 mph, 45 mph, 50 mph, 55 mph, and 60 mph were removed from the dataset when calculating the variance. The resulting speed variances from these trimming processes were then compared. At all sites, the 15th percentile speed was about 65 mph, 25 mph above the posted minimum speed of 40 mph.
The results showed no discernable contribution to speed variance for vehicles with a speed of less than 55 mph, primarily because very few vehicles at each site traveled at speeds less than 55 mph. In fact, at each site vehicles with speeds under 55 mph made up 1% of those recorded, while the percentage of vehicles with speeds of less than 40 mph was negligible (i.e., 0.15%). Although the contribution to the standard deviation of vehicles with speeds less than 55 mph is very minor, the safety implications of the presence of vehicles with very low speeds cannot be ignored. Even though only a few vehicles cause speed differential conflicts, these vehicles could be a contributory factor in crashes.
Highway travel is generally composed of free-flowing and platooned vehicles. In free-flowing traffic, drivers can choose their speeds as they desire as long as conditions are such that slow-moving vehicles do not impede their ability to change lanes at will. Platooned vehicles travel close to each other mostly because of lack of passing opportunities, thus causing other vehicles to be trapped behind the lead vehicle. No definition exists in the literature of a headway value below which vehicles are considered to be moving in a platoon. Thus, in this study, four definitions were considered-less or equal to 1, 2, 3, and 4 seconds.
The analysis showed that six-lane highway sections carried larger proportions of platooned vehicles than four-lane sections. Further, the middle lanes of six-lane sections carried more platoons than the shoulder and median lanes. To study the effect of platooned vehicles on the distribution of speed, the mean speeds of platooned vehicles were compared with the mean speeds of nonplatooned (or free-flowing) vehicles. The statistical analysis here uses a t-test in which platooned and nonplatooned vehicles were paired by site and by lane of travel. The results showed that the difference between the speeds of platooned and nonplatooned vehicles were insignificant for both four- and six-lane highway sections regardless of whether the cut-off point was 1, 2, 3, or 4 seconds of time headway. These results indicate that platooned vehicles are not slow moving and thus do not create a need for free-flowing vehicles catching up behind them to pass. However, it should again be noted that the highway sections analyzed were relatively uncongested, operating at levels of service B or better for a majority of the hours in a year.
BEFORE AND AFTER COMPARISON
To understand the change in speed characteristics following the increase in the speed limit, table 5 presents a comparison of before-and-after data. In 1996, the speed limit was 65 mph at all the sites indicated in the table. Data-collection sites for both 1996 and 2002 were physically very close, and the field review of the sites indicated that for all practical purposes the geometric characteristics prevailing at these sites would produce similar driver behavior.
The results in table 5 show that the average speeds across all sites increased by 5 mph to 72 mph. The 15th percentile speed also showed a significant increase of 3 mph when averaged across all sites (p ≤ 0.0001). A statistical F-test comparison of the variances indicated no significant difference between the 1996 and 2002 data (p = 0.50). However, significant differences were found in the variances on four-lane sections (p = 0.0003). Further analysis indicated that in 1996, the average speed on six-lane sections was 4.75 standard deviations above the 40 mph minimum posted speed limit. In 2002, it was 5 standard deviations above the 40 mph minimum. In four-lane sections, the results show that the average speeds were 6 and 5 standard deviations above 40 mph in 1996 and 2002, respectively. Examination of the coefficients of variation between the two datasets indicated that 2002 data show significant large variations compared with 1996. However, the coefficients of variation are still below 10%, indicating a reasonable equity in travel speeds.
DISCUSSION OF RESULTS
This paper presents a review of traffic operating characteristics on rural interstate highways in Florida. Using various analytical techniques, we determined speed characteristics in relation to the posted minimum speed limit of 40 mph. Our intent was to examine the relevance of the 40 mph minimum speed limit in light of the increase in the maximum speed from 65 mph to 70 mph.
It is clear from the analysis that raising the speed limit increased average speeds on rural interstate highways. The comparison of 1996 data with 2002 data showed that average speeds rose by 5 mph, which is the same amount of the speed limit increase. The comparison further showed a slight increase in the coefficient of variation after the maximum speed went up; however, the increase is statistically insignificant and under 10%, a threshold that can be considered to indicate uniform operations. In addition, the 15th percentile speed showed an increase of 3 mph when averaged across all sites. In relation to the 40 mph posted minimum speed, the 2002 average speed on all sections was 5 standard deviations above this minimum speed, compared with 5.4 standard deviations for the 1996 data.
In light of the above data and analyses, from a traffic operations standpoint, several questions arise: Is the practice of posting the 40 mph minimum speed irrelevant or is it successful in ensuring that vehicles do not travel below 40 mph? Should the 40 mph posted minimum speed limit be scrapped or should it be raised to a higher value? What should that value be? These are important questions that could not be adequately answered through the research paradigm reported here. However, the data reveal a few pointers.
First, the 40 mph posted minimum speed limit probably does not have a significant influence on driver behavior given that the number of vehicles traveling below 55 mph at all sites was negligible (i.e., 1%). If these signs influenced drivers, we would expect a higher percentage of vehicles to travel at speeds in the 40 mph to 50 mph range, as is the case on the higher side of the speed distribution where a large percentage of drivers maintain speeds between 70 mph and 80 mph.
It has been suggested in the past (e.g., McShane et al. 1998) that the 15th percentile speed may be used as a measure of the minimum reasonable speed for the traffic stream. (This suggestion mirrors the attempt to use the 85th percentile speed as a measure for setting the maximum speed limit). The data reported here indicate that, in all sections studied, the 15th percentile speeds on the aggregate ranged from 60 mph to 70 mph, which is 20 mph to 30 mph above the posted minimum speed limit value. Does this mean that the minimum speed limit should be set at 60 mph? There are number of concerns that would need to be addressed before a change like this could be made. First, Florida statutes (Florida Statutes 2002) state that "no school bus shall exceed the posted speed limit or 55 mph." Second, as a tourist state, some Florida visitors drive recreational vehicles (sometimes towing a trailer) or motor homes, and field review indicated that these are the vehicles that tend to make up the lowest 15% of the speed distribution at all sites. Third, a safety analysis would be needed to fully justify any change in the minimum highway speed.
Instead of increasing the minimum speed, should it be eliminated? After all, the results of a survey conducted as part of this research showed that 25 states do not post minimum speeds on interstate highways. Currently, Florida statutes state that: "The minimum speed limit on interstate and Defense Highways, with at least 4 lanes, is 40 mph." The Florida Highway Patrol in the context of this research study indicated that such a statute is required to enable law officers to issue citations. A question was raised that in the absence of the minimum speed rule, can the law officers use another Florida statute that states "No person shall drive a motor vehicle at such a slow speed as to impede or block the normal and reasonable movement of traffic" to warn or issue citations to slow moving vehicles? One police officer pointed out that if a vehicle is alone on the highway traveling at, say 25 mph, what traffic is the driver impeding?
Further research is needed to ascertain the effect of the current posted minimum speed limit on driver behavior. While the data seem to indicate that the 40 mph minimum speed might not be that relevant based on prevailing operating speed distributions, it is not clear what the effect would be if the signs were removed from rural interstate highways. The answer to most of the questions raised above requires field evaluation, as simulation analysis would not appropriately depict driver behavior on roadways with and without posted minimum speed limit signs.
Additional research that is planned includes collecting data on interstate highway sections in states that do not have minimum speed limits posted but have similar geometric and driver characteristics. A comparison of multistate data might shed some light on the relevance of posting minimum speed limit signs. Multistate data would also be of interest to traffic engineers who want to compare safety characteristics on sites with and without posted minimum speed limits.
2002 Florida Statutes. Title XXIII, Chapter 316, section 183(3).
Hauer, E. 1971. Accidents, Overtaking, and Speed Control. Accident Analysis and Prevention 3:1-13.
Lave, C. 1985. Speeding, Coordination, and the 55 mph Limit. American Economic Review 75:1159-1164.
McShane, W.R., R.P. Roess, and E.S. Prassas. 1998. Traffic Engineering. Upper Saddle River, NJ: Prentice-Hall.
Mussa, R. 2003. Nationwide Survey of the Practice of Posting Minimum Speed Limit Signs on Interstate Highways, manuscript.
National Committee on Uniform Traffic Laws and Ordinances. 1954. Uniform Vehicle Code. Washington, DC.
______. 1964. A Comparative Survey Based on the Uniform Vehicle Code. Traffic Laws Annual 1.
Transportation Research Board (TRB). 1984. Special Report 204: 55: A Decade of Experience. Washington, DC: National Research Council.
U.S. Department of Transportation (USDOT), Federal Highway Administration. 2000. Highway Capacity Manual. Washington, DC.
West, L.B. and J.W. Dunn. 1971. Accidents, Speed Deviation and Speed Limits. Traffic Engineering 41:52-55.
1LOS classifies the quality of operation provided by the roadway from A through F, with "A" representing the most favorable driving conditions and "F" the worst, measured at the peak hour period of the day (USDOT 2000).
ADDRESSES FOR CORRESPONDENCE
1 Corresponding author: V. Muchuruza, Department of Civil Engineering and Environmental Engineering, Florida A&M University-Florida State University, 2525 Pottsdamer Street, Tallahassee, FL 32310. E-mail: email@example.com
2 R. Mussa, Department of Civil Engineering and Environmental Engineering, Florida A&M University-Florida State University, 2525 Pottsdamer Street, Tallahassee, FL 32310. E-mail: mailto:firstname.lastname@example.org