Transportation patterns have changed a lot lately. USDOT data shows fewer Americans aged 16-19 now have driver's licenses compared to 1998. Many cities like Boston, Chicago, and San Francisco have set bold targets to boost non-motorized transport. Yet federally funded projects still need traffic projection tools that assume 1-2% yearly vehicle growth for 10-30 years ahead. This mismatch shows why transportation professionals must become skilled at different traffic growth forecasting methods - whether exponential, linear, or logistic.
Your choice of yearly traffic growth rate formula affects infrastructure decisions by a lot. To name just one example, studies show new roadway capacity gets absorbed by induced traffic within three years - anywhere from 50% to 100%. Picking the right forecasting approach for your specific case helps avoid pricey overbuilding while meeting real transportation needs.
Traffic forecasting is the life-blood of transportation infrastructure planning worldwide. Over the last several years, forecasting methods have changed by a lot, and they still determine how we spend billions on infrastructure each year [1]. The stakes for getting traffic growth projections right have never been higher as we head into 2025.
Getting traffic forecasts right is crucial to make critical project decisions about structural and geometric capacity, financial viability, and environmental effects [2]. Transportation agencies can't properly plan for future needs or justify infrastructure investments without reliable traffic projections. These growth estimates also directly shape air quality assessments, noise studies, and energy consumption calculations – which change drastically based on expected traffic levels [2].
Transportation planners usually create growth projections that look 20 years past a project's expected opening date [2]. Plans that need more refinement stretch this timeline to 25-30 years. This helps extend a plan's useful life, especially if project development doesn't start right away [2]. The 30-year mark tends to be the limit for reasonable accuracy, as detailed analysis becomes less reliable beyond that point [2].
Getting traffic forecasts right has become even more critical in 2025. These forecasts try to figure out future roadway use in communities, which helps evaluate transportation facilities and land use plans better [3]. Planners can use these numbers to figure out if they should build new roads or add more transportation services [3].
Planning-level forecasts depend on future average annual daily traffic (AADT) volumes. Various traffic analysis tools generate these numbers, and the area's socioeconomic characteristics play a big role in shaping these projected volumes [3]. These projections then help determine geometric design values for roadway designs, including how intersections should be configured and where to put continuous traffic count sites [3].
The rise of automobile technology, especially self-driving vehicles and electric cars, has become a key factor that shapes future traffic patterns [3]. Getting traffic predictions right has become essential to move vehicles efficiently, cut down on congestion, and create the best possible routes [3].
Wrong traffic forecasts pose big risks to infrastructure planning and financing. Research on toll road traffic shows that forecasts often miss the mark and usually aim too high [3]. This consistent bias leads to serious problems:
Financial Distress: Traffic volumes that fall short of optimistic forecasts have led many toll road public-private partnerships (PPPs) to fail. This results in high-profile project troubles, bankruptcies, renegotiations, and government bailouts [4].
Resource Misallocation: Overestimated traffic growth can lead agencies to pour billions into unnecessary road expansions while neglecting maintenance or alternative transportation options [1].
Self-Fulfilling Prophecies: Traffic growth often works both ways – the expansion itself makes projections come true through induced demand [1]. A well-known 2009 study by Duranton and Turner found that new urban freeways create new traffic at an almost perfect 1-to-1 ratio [1].
The biggest problem with traffic forecasting comes from three main sources: error, uncertainty, and bias [4]. Error (honest human mistakes) and uncertainty (unexpected external changes) should balance out over time. But real-world evidence shows traffic forecasts consistently aim too high, which points to built-in biases in the forecasting process [4].
These biases show up at the time forecasts get inflated to achieve specific goals – like a bidder trying to win a project or government officials seeking project approval [4]. Sometimes this intentional inflation has landed forecasting firms in legal trouble [4].
Traffic modeling's real purpose isn't always accuracy – it's often about creating believable stories to justify more construction [5]. This flawed approach leads to predictions of increasing traffic even on roads where volumes have been dropping for years [5].
The National Academy of Sciences has noticed this issue. They concluded that current travel demand forecasting models "do not offer the national or regional-level prediction capabilities needed to assess system level impacts from Interstate investments" [5]. All the same, cities and states keep spending huge amounts on increasingly complex models that often fail to predict traffic accurately [5].
The most worrying part is how traffic projections often ignore induced demand principles. This creates a circular argument: "New freeways will be necessary to accommodate all the traffic generated by our new freeways" [1]. This self-reinforcing cycle traps communities into constantly expanding infrastructure while missing chances to create greener transportation solutions.
Traffic forecasting needs reliable baseline data and standardized methods. Transportation engineers use specific traffic metrics, selected timeframes, and standardized adjustment factors. These elements help create projections that stand up to scrutiny. These foundations are the life-blood to get valid traffic growth rate formulas.
Annual Average Daily Traffic (AADT) is the life-blood metric for traffic forecasting calculations. AADT shows the mean traffic volume across all days for a year at a specific location along a roadway [6]. Simple daily traffic counts don't match AADT, which represents data for the entire year and provides more detailed measurement [6].
Transportation agencies use several methods to calculate AADT:
AADT shows vehicle traffic load on road segments and serves as a key input parameter for transportation planning and fund allocation [6]. AADT creates the baseline for traffic growth projections.
Design Hour Volume (DHV) shows the traffic volume expected during the design hour of the design year. DHV matches the 30th highest hourly volume of the year [6]. Roads must handle peak traffic conditions rather than average flows. The K-factor expresses the relationship between DHV and AADT: DHV = K × AADT [7].
Timeframe selection shapes traffic projection outcomes. The base year matches the projected opening date to traffic [8]. The design year shows the future point for traffic volume forecasts.
Transportation planners use a 20-30 year horizon for long-range planning [9]. They think over intermediate forecast years—including the project's opening date [9]. These points help confirm projection accuracy as the project moves forward.
Planning activities need a minimum 20-year horizon to evaluate transportation needs and solutions [5]. Refinement plans often stretch to 25-30 years. This increases the plan's useful life, especially when project development takes time [5].
Detailed analysis limits extend to about 30 years. Anything beyond becomes approximation rather than prediction [5]. Travel demand models work best when extrapolation stays within five years of the model's future year [5].
Standardized factors normalize traffic data for better comparisons and projections. Two key normalization metrics include:
K-factor: Shows the proportion of AADT during the design hour (typically the 30th highest hourly traffic volume of the year) [7]. Engineers use the K-factor to convert daily traffic volumes into peak-hour designs that handle realistic worst-case scenarios.
D-factor: Shows the proportion of design hourly volume in the heavier direction, also called the Directional Split [7]. D represents directional imbalance during peak hours and must exceed 0.5 (50%) [7].
These factors work together through these relationships:
The Directional Design Hourly Volume (DDHV) drives many geometric design decisions. Roadway capacity must handle this peak directional flow [7]. Getting accurate K and D values needs a traffic profile group (TPG) for the analyzed roadway segment or link [7].
These baseline metrics and normalization factors help transportation engineers build the foundation for various traffic growth rate formulas—whether exponential, linear, or logistic. The quality of baseline inputs determines how reliable the traffic projections will be.
Image Source: ResearchGate
Traffic forecasting relies heavily on exponential growth. This mathematical concept helps model how vehicle volumes speed up under specific conditions. Traffic engineers need to know which models work best for different growth patterns.
The exponential traffic growth rate formula helps calculate future traffic volumes. It works by looking at percentage increases from previous years. You'll see this mostly in new areas that have plenty of land and road space. Here's the formula:
VolumeFY = VolumeBY × (1 + r)^(FY-BY)
Where:
Traffic volume grows by the same percentage each year with this formula. The traffic volume curve starts slow but speeds up without any limits as time goes on. A steady 2% yearly growth rate (r = 0.02) means traffic volume doubles in about 35 years [10]. This is a big deal as it means that road capacity needs might get too high for realistic infrastructure.
You should use exponential growth models in these specific traffic forecasting cases:
Traffic engineers often use exponential growth models at the start of a long-term growth curve. A complete model might combine three stages: exponential growth for the first 5 years, linear growth for the next 10 years, and declining growth for the final 5 years [11].
This matches how populations grow naturally. Unrestricted populations start with exponential growth until they hit limits like environmental factors, available resources, or space [2].
Exponential traffic growth models work well in specific cases, but they have major limitations:
Overestimation risk: Long-term exponential curves can seriously overstate future traffic volumes [11]. This happens because compound growth assumes unlimited expansion, but real-life traffic faces capacity limits.
Inaccuracy in urban settings: Standard forecasting models often predict too much traffic growth on city streets [3]. Data shows that actual traffic volumes usually end up about 6% lower than predicted [12].
Unrealistic behavior assumptions: These models think roads can handle more than their maximum capacity, which isn't possible. No road can carry more than 100% of its capacity, yet you'll hear claims like "without this expansion, the roadway will be at 110% capacity" [3]. Models wrongly assume people will keep using packed roads no matter what.
Failure to account for behavioral adaptation: People adjust their habits as roads get crowded. They find new routes, travel at different times, use other transportation, or pick closer destinations [3]. Standard traffic models don't account for these behavioral changes.
Environmental and infrastructure constraints: Real-life exponential traffic growth eventually levels off as networks reach their limits. Research shows that "growth rates have to be large enough over time to be affected by the capacity value so the curve will flatten out" [11].
Traffic planners should switch from exponential to linear or logistic models after five years. These better reflect real-life constraints. Most new roads see steady traffic increases for the first 10 years (5% to 10% compound growth). The rate then drops to 1% to 3% over the next 30 years [2].
Exponential growth formulas help forecast short-term traffic, especially for new development in unrestricted areas. In spite of that, you need to think over the timeframe and context carefully to avoid getting pricey overdesigned transportation infrastructure.
Linear traffic growth shows a more steady approach to traffic forecasting compared to its exponential counterpart. It keeps a constant rate of increase over time. Transportation engineers prefer linear projections because they are predictable and realistic, especially for mature roadways or those with physical limits.
You can calculate future traffic volumes using this simple equation:
VolumeFY = VolumeBY × (1 + r × N)
Where:
The formula adds the same number of vehicles each year instead of compounding them. You can express the relationship between growth rate and growth factor as:
Growth Factor (GF) = 1 + (Growth Rate × Number of Years)
Growth Rate = (Growth Factor - 1) / Number of Years
To name just one example, see a 20-year growth factor of 1.63 (traffic increased from 19,600 to 32,000 vehicles). The annual linear growth rate would be (1.63 - 1.0) / 20 = 0.032 or 3.2% [4].
Linear traffic growth models work best in these specific cases:
Transportation planners often choose linear growth projections because "often there is insufficient data to support use of a more specific type of curve" [4]. This practical approach recognizes data limitations while providing applicable forecasts.
The main difference between linear and exponential growth lies in how increases build up over time. Linear growth maintains a steady rate of change—traffic volume increases consistently with time [1]. Exponential growth happens relative to current volume, which leads to faster acceleration [14].
Short-term forecasts might show small differences between linear and exponential projections. These gaps grow significantly over longer periods. A road segment with 1,000 vehicles would reach 1,600 vehicles after 20 years using linear growth with a 3% annual rate. The same scenario with exponential growth would predict about 1,806 vehicles—a 13% difference [15].
Linear growth has several benefits over exponential models:
Linear modeling fits well with traffic patterns in developed areas where growth happens through fixed steps rather than percentage increases. Transportation planners who balance limited resources against future needs often find linear growth formulas most useful for mid-range forecasting.
Analysts should look at both historical patterns and predicted development conditions when choosing between linear and exponential models. Many transportation agencies note that "linear growth is used since often there is insufficient data to support use of a more specific type of curve" [4]. This makes it a practical choice for many real-life applications.
Traditional growth formulas don't work well when traffic networks reach their maximum capacity. More sophisticated modeling approaches become necessary. Logistic curves capture how traffic growth slows down as it approaches a network's physical limits, unlike exponential and linear models.
The logistic growth equation calculates traffic volume based on the network's carrying capacity:
V = VC / (1 + ((VC - V0)/V0) × e^(-rt))
Where:
This formula shows how traffic increases almost exponentially when it's well below capacity. The growth gradually slows as it approaches capacity and flattens at the saturation point [16]. Logistic curves model natural resistance that increases as volume approaches carrying capacity. This prevents growth beyond physical limits [16].
Link capacity constraints often cause congestion in urban transportation networks. Physical queuing increases travel costs by a lot [17]. Traditional static traffic assignment (STA) models have notable limitations. They don't deal very well with explicit capacity constraints, which leads to errors around bottlenecks and inaccurate travel time projections [5].
Capacity-constrained traffic assignment models store excess vehicles in residual queues to address these shortcomings. These models acknowledge that all vehicles may not reach their destinations within a given time period [5]. This approach helps especially when you have congested networks where traffic demands exceed capacity during peak periods [5].
To name just one example, typical signal synchronization strategies don't perform well in saturated high-density grid road networks (HGRN) [18]. Traffic modeling must include elements like long green and long red (LGLR) signal timing. This helps maintain continuous traffic flows while keeping queue formation to a minimum [18].
Saturation flow rate forms the foundations of capacity calculations. It represents the maximum number of vehicles that can traverse a point during ideal conditions. A modified version of the Highway Capacity Manual equation determines this rate [19]:
s = so × N × fw × fHV × fg × fp × fbb × fa × fLU × fLT × fRT × fLpb × fRpb × PHF
Factors like lane width (fw), heavy vehicles (fHV), and peak hour factor (PHF) adjust the base saturation flow rate (so) [19].
Traffic delays and queues serve as key performance measures to determine intersection level of service (LOS) [9]. Yes, it is these metrics that help evaluate lane length adequacy and estimate fuel consumption and emissions [9].
Traditional steady-state models become inadequate as traffic flow approaches capacity because stochastic equilibrium cannot be achieved [9]. Analysis of signals in saturated conditions must consider both deterministic (uniform) and stochastic (random) components [9].
The critical density point occurs at about half the carrying capacity. This is where traffic flow maximizes before deteriorating [16]. The increasing rate of the logistic curve begins to decline at this inflection point. This signals the shift from free-flowing traffic to congested conditions [16].
Traffic growth formulas work best when planners understand how trips flow through transportation networks. Trip generation and assignment models are vital inputs for traffic forecasting. These models help planners convert land use changes into actual network flows.
The Institute of Transportation Engineers (ITE) Trip Generation Manual is the go-to resource for measuring traffic impact in North America. The manual started in 1976 with just 50 land uses in 200 pages. Now, the 11th Edition has grown to cover 179 land uses across more than 4,500 pages [20]. This complete resource helps transportation professionals estimate trips from specific land developments without getting pricey original research for each project.
The ITETripGen app, which comes with the 11th Edition, boosts functionality by letting users:
Trip distribution matches origins with destinations after trips are generated. This second step in the conventional four-step transportation planning process creates a matrix showing trips between locations [21].
The gravity model is the main way to distribute trips. It works like Newton's Law of Gravity. The model shows that trips between two zones depend on their attraction values (usually employment or retail activity) and the travel difficulty between them [22].
The math behind this uses "friction factors" to show how hard it is to travel between zones. These factors change based on trip purpose since people travel differently for work, shopping, or fun [22].
Route selection between origins and destinations happens in the final step. This trip assignment shows how people pick their paths through a transportation network [23].
Modern assignment models do more than find the shortest path - they look at capacity limits and balance principles. The User Equilibrium (UE) approach follows Wardrop's principle. It assumes travelers pick routes that save them time while everyone else does the same [23]. When roads get busy, speeds drop and people find other routes [24].
Busy networks need capacity restraint algorithms. These tools adjust travel speeds based on volume-to-capacity ratios until they find the right balance [24]. The results show traffic volumes that help create accurate growth projections.
A successful traffic forecast depends on strong validation and calibration. Transportation agencies can improve their future methodologies by comparing their projected volumes with actual post-construction counts.
Studies show measured traffic runs about 6% below forecasted volumes, with predictions deviating by an average of 17% [7]. The accuracy varies based on facility types. Roads with higher volumes, better functional classifications, and shorter time spans show better forecast precision. Travel models also help improve the accuracy [7].
The forecast accuracy has evolved over time. Recent forecasts show better accuracy compared to older ones, though the deviation patterns have changed through decades [7]. Traffic exceeded predictions for projects from the 1980s through early 2000s. However, recent projects show traffic volumes below the forecasted numbers [7].
Accurate traffic forecasting relies on calibration - the process of adjusting model parameters to better reflect local driver behavior [25]. Analysts need to review hundreds of model parameters that affect simulation results in connected ways [25]. The process follows three main steps:
The goal of calibration is to find model parameters that minimize the mean squared error between model outputs and field measurements [25].
Sensitivity testing reveals how changing basic assumptions affects projections. Research shows that dropping the projected annual VMT growth from 1.2% to 0.9% would lower average annual investment by 7.2% for maintenance and 8.1% for improvement scenarios [6].
Unemployment rates have a big effect on accuracy. Traffic would be 1% above forecast instead of 6% below if adjusted for higher unemployment during post-recession years (2008-2014) [7].
Transportation agencies prepare multiple forecast scenarios based on pessimistic (Low Case), best-estimate (Base Case), and optimistic (High Case) assumptions [26]. Traffic forecasts work better when viewed as a range of possible outcomes rather than a single expected number [7].
Traditional traffic growth rate formulas have basic flaws that lead to wasted resources and poor transportation planning. Studies show these old limitations make them unreliable for modern infrastructure decisions.
Real-world data shows traditional forecasting methods overestimate traffic growth, especially in urban arterials and corridors. Traffic volumes after construction usually end up 6% lower than forecasted numbers, and average deviations reach 17% from predictions [7]. This bias toward higher estimates is most noticeable in developed urban areas.
Rail project forecasts show even bigger errors. Passenger numbers were too high in 72% of rail projects studied—off by an average of 106% [27]. Road forecasts, while not as extreme, still show concerning differences. Half of all road projects had a striking 720% gap between actual and predicted traffic [27].
Many traffic models' basic theory misses induced demand—the fact that bigger roads create new trips that wouldn't happen otherwise [28]. This explains why expanded roads quickly fill up with traffic again.
Studies show new road capacity gets filled by induced traffic within three years—anywhere from 50-100% [10]. Transportation agencies leave this effect out of their models [27]. Their projections then underestimate how congested expanded roads will become.
Today's transportation policies focus on walking, biking, and transit use. They often set specific targets to reduce single-driver car trips [10]. In spite of that, traffic models work separately from these policy goals. They stick to the old idea of "fixing" congestion by making roads bigger [11].
This gap shows up when agencies spend billions on unnecessary road expansions while maintenance and alternative transport get underfunded [11]. On top of that, these models don't account for basic changes in how people travel, including demographic shifts and social patterns that post-construction studies have found [27].
Traffic growth forecasting shapes how we plan, build, and maintain transportation infrastructure. Our analysis explored different mathematical models—exponential, linear, and logistic—that serve distinct scenarios in traffic projection. Notwithstanding that, each approach has important limitations that transportation professionals must recognize.
Clear evidence shows traditional forecasting methods consistently overestimate traffic volumes, especially when you have urban settings. Transportation agencies have wasted billions in potentially misallocated infrastructure spending due to this systematic bias. Then, these agencies must fine-tune their approaches to include emerging travel patterns and behavioral changes.
Without doubt, the gap between forecasting practices and today's transportation policy goals presents one of our biggest challenges. Many 3-year old cities have set ambitious targets to increase non-motorized transportation modes. Their traffic projections often use outdated assumptions of continuous vehicle volume growth. This mismatch weakens efforts to create greener, multimodal transportation systems.
Projection accuracy suffers from inadequate consideration of induced demand. Research shows new roadway capacity becomes 50-100% filled with induced traffic within just three years. Standard modeling procedures still largely exclude this phenomenon.
Traffic forecasting must grow beyond simple mathematical formulas as we look toward 2025 and beyond. This growth needs actual post-construction validation data. It should apply appropriate growth models based on network maturity and line up projections with broader policy objectives. The future of effective infrastructure planning ended up depending not on perfecting a single universal formula. Instead, it relies on choosing the right forecasting methodology for specific contexts while acknowledging inherent uncertainties.
Q1. What is the most accurate method for forecasting traffic growth? There is no single most accurate method, as different approaches suit different scenarios. For short-term forecasts in developing areas, exponential growth models work well. Linear models are better for mature corridors with steady growth. For networks approaching capacity, logistic growth curves are most appropriate. The key is selecting the right model based on the specific context and validating projections against real-world data.
Q2. How do transportation planners account for induced demand in traffic forecasts? Many traditional forecasting models fail to adequately account for induced demand. Research shows that 50-100% of new road capacity is often absorbed by induced traffic within three years. Advanced models are starting to incorporate this phenomenon by adjusting projections based on capacity increases. However, many agencies still exclude induced demand, leading to underestimation of post-expansion congestion.
Q3. What data sources are used to establish baseline traffic volumes for projections? The primary metrics used are Annual Average Daily Traffic (AADT) and Design Hour Volume (DHV). AADT represents the mean traffic volume across all days of the year at a specific location. DHV typically corresponds to the 30th highest hourly volume of the year. These baseline figures are then adjusted using factors like the K-factor (proportion of AADT during peak hour) and D-factor (directional split) for more accurate projections.
Q4. How far into the future can traffic forecasts reliably predict volumes? Most transportation agencies use a 20-30 year horizon for long-range planning. However, the reliability of forecasts decreases significantly beyond about 5-10 years. Detailed analysis is generally limited to about 30 years, with anything beyond that considered approximation rather than prediction. For travel demand models, it's recommended to limit extrapolation to five years beyond the model's future year.
Q5. How do changes in transportation technology impact traffic forecasting? Emerging technologies like autonomous vehicles and electric cars are significantly influencing future traffic demand trends. These advancements can affect travel behavior, road capacity, and congestion patterns. While traditional forecasting models often struggle to account for these changes, more advanced approaches are beginning to incorporate technological factors. Accurate long-term forecasting increasingly requires considering how these innovations might reshape transportation networks and travel patterns.
[1] - https://study.com/academy/lesson/comparing-linear-exponential-functions.html
[2] - https://railknowledgebank.com/Presto/content/GetDoc.axd?ctID=MjE1ZTI4YzctZjc1YS00MzQ4LTkyY2UtMDJmNTgxYjg2ZDA5&rID=MzI0NA==&pID=MTQ3Ng==&attchmnt=True&uSesDM=False&rIdx=MTM1Nzg=&rCFU=
[3] - https://t4america.org/2023/06/29/the-traffic-forecast-used-to-justify-your-road-widening-is-bogus/
[4] - https://www.oregon.gov/odot/Planning/Documents/APMv2_Ch6.pdf
[5] - https://www.researchgate.net/publication/245306516_Capacity-Constrained_Traffic_Assignment_in_Networks_with_Residual_Queues
[6] - https://www.fhwa.dot.gov/policy/24cpr/pdf/Chapter9.pdf
[7] - https://link.springer.com/article/10.1007/s11116-021-10182-8
[8] - https://www.itetripgen.org/
[9] - https://www.fhwa.dot.gov/publications/research/operations/tft/chap9.pdf
[10] - https://nacto.org/publication/urban-street-design-guide/design-controls/design-year/
[11] - https://t4america.org/community-connectors/what-they-mean/induced-demand/
[12] - https://www.kittelson.com/ideas/myth-busters-are-long-range-traffic-forecasts-the-best-tool-to-guide-decisions-about-our-streets/
[13] - https://www.vdot.virginia.gov/media/vdotvirginiagov/doing-business/technical-guidance-and-support/technical-guidance-documents/transportation-and-mobility/traffic-forecasting-guidebook-v1.1-2024-05_acc.pdf
[14] - https://www.statsig.com/perspectives/exponential-vs-linear-growth
[15] - https://www.tmr.qld.gov.au/-/media/busind/techstdpubs/Project-delivery-and-maintenance/Cost-benefit-analysis-manual/42Volumecapacityratio.pdf?la=en
[16] - https://www.sciencedirect.com/topics/computer-science/logistic-growth
[17] - https://www.sciencedirect.com/science/article/abs/pii/S1366554524000760
[18] - https://pmc.ncbi.nlm.nih.gov/articles/PMC4309220/
[19] - https://www.fhwa.dot.gov/ohim/hpmsmanl/appn5.cfm
[20] - https://www.ite.org/ITEORG/assets/File/ITEJ%20Published/2021/ITE_ITE_October2021.pdf
[21] - https://eng.libretexts.org/Bookshelves/Civil_Engineering/Fundamentals_of_Transportation/03%3A_Modeling_Methods/3.04%3A_Trip_Generation
[22] - https://uta.pressbooks.pub/oertransportlanduse/chapter/chapter-11-second-step-of-four-step-modeling-trip-distribution/
[23] - https://uta.pressbooks.pub/oertransportlanduse/chapter/chapter-13-last-step-of-four-step-modeling-trip-assignment-models/
[24] - https://www.kyovaipc.org/modeltraffassign.htm
[25] - https://ops.fhwa.dot.gov/trafficanalysistools/tat_vol3/sect5.htm
[26] - https://ppp.worldbank.org/public-private-partnership/sites/default/files/2022-06/WB_PPIAF_UnderstandingTrafficRisk_IssueBrief_crops_final.pdf
[27] - https://nacto.org/wp-content/uploads/2015/04/post_construction_evaluation_of_traffic_forecast_accuracy_parthasarathi.pdf
[28] - https://www.strongtowns.org/journal/2022/4/4/ignoring-induced-demand-is-engineering-malpractice
Safer FMCSA provides accurate, up-to-date trucking carrier data across the USA, specializing in flatbeds, reefers, dry vans, step decks, and power-only carriers.
Copyright © 2024 SAFER FMCSA By UBAID & SAMI. All Rights Reserved.