Projection models for road transport demand: impact and innovation


Demand projection is a crucial tool in planning and managing transportation infrastructure. It makes it possible to predict future mobility needs, enabling informed decision-making and the efficient allocation of resources. In the Brazilian context, especially involving road concessions, projection models play a fundamental role, directly influencing public policies and private investments.

Demand projection models are essential for several activities related to road transportation, such as:

  1. Infrastructure planning: They help identify the need for expansion, improvement or construction of new roads, toll plazas, accesses, among others.
  2. Economic Feasibility Assessment: Allows the financial viability of infrastructure projects to be assessed, influencing public and private investment decisions.
  3. Traffic Management: Facilitates the appropriate dimensioning of capacities and the implementation of traffic control measures, with a view to improving road safety and fluidity.
  4. Urban and Regional Planning: They contribute to urban and regional development by predicting the impacts of road transport on the highway’s surrounding areas.
  5. Sustainability: Allows environmental and social variables to be considered when making decisions, seeking to minimize negative impacts and promote sustainability in the transport sector.

The transportation structure is fundamental to the nation’s economic progress and prosperity, as it facilitates the connection between different regions, allowing the flow of natural resources, manufactured products, consumer goods and people both within the national territory and across borders. In Brazil, there is a considerable demand for transportation infrastructure resources, especially for highways, where approximately 60% of the country’s cargo is transported. The lack of public resources to maintain and expand this system has resulted in the concession of highways to the private sector.

Recently, a more rigorous stance has been observed in the analysis and validation of demand projection studies, partly due to past experiences of concessions that faced challenges related to underestimating or overestimating demand. An example of this was the 3rd round of the Federal Highway Concessions Program (PROCROFE). According to Stefanello (2022), the concessionaires under study had, on average, a -17.94% variation in volume compared to what was projected in the traffic studies. If the six highways granted at this stage presented the volume estimated in the traffic studies, they would have R$2.86 billion in additional revenue to what was accounted for between 2014 and 2021. Thus, with traffic volumes below expectations, 5 of the 6 concessions were returned to the granting authority.

Faced with this scenario, added to the economic instability of the last decade and the Covid-19 pandemic, both regulatory institutions and concession companies have shown themselves to be more careful when assessing feasibility studies, demanding robust and transparent methodologies.

Chapter 8 of the Traffic Studies Manual (DNIT, 2006) does not establish a single methodology for forecasting demand but offers various approaches and recommends the adoption of more comprehensive models that consider multiple socio-economic variables. Despite this flexibility, the Gross Domestic Product has been predominantly used as the sole indicator of road demand, mainly due to the availability of reliable historical series for correlation analyses and official projections of this variable.

However, in addition to factors such as harvest periods in agricultural regions like the central-west, there are other variables that can directly influence road demand. At times, both nationally and regionally, there has been a disconnect between the GDP variable and traffic, due to various factors. An example of this was observed during the Covid-19 pandemic, when some concessionaires recorded stable or even growing demand for heavy vehicles, despite the adverse economic context.

In addition, factors such as the rate of motorization, changes in the dynamics of the mobility of people and goods, and logistics processes also play significant roles in road demand. This indicates that the economy should not be the only factor determining traffic.

In this context, SYSTRA has stood out for its innovative approach to developing demand projection models. Our consulting team has dedicated significant efforts to researching and developing new methodologies, considering a wide range of socio-economic and contextual variables, according to the specific characteristics of each highway under study.

Within the context of regionalization, the region’s infrastructure development scenarios are also considered, such as the forecast of modal migration with the expansion of the rail network and new integration logistics terminals.

One of SYSTRA’s main strategies is the adoption of advanced statistical analysis, making intensive use of the Statsmodels library in the Python programming language. This approach makes it possible to analyze various predictive variables and choose those that best suit the circumstances of the regression for estimating future demand. This approach allows for faster, more accurate and flexible modeling, as well as facilitating integration with other tools and geographic information systems (GIS).

There has also been a methodological evolution when it comes to analyzing highways that include urban crossings or bypasses. These stretches play a crucial role not only as logistical connections between different regions, but also in mobility within municipalities. Among the most recent challenges we face is assessing the uncertainties that affect commuting, such as demographic and land use changes, population reduction or increased urban density, as well as changes in consumption patterns, like the growth of e-commerce. These factors can not only influence the volume of trips, but also the choice of transportation modes in the region.

Another improvement is the incorporation of telephony data into our demand studies, a practice that has become routine in our analyses. In addition, SYSTRA has sought inspiration from the literature and practices of other countries, adapting and improving internationally recognized methodologies for the Brazilian context. This includes accounting for cultural, economic, and regulatory aspects specific to the region studied, guaranteeing the relevance and effectiveness of the models developed.

One of SYSTRA’s strengths is its ability to offer clients a variety of demand scenarios, ranging from the most optimistic to the most conservative. This allows for a more comprehensive analysis of the risks and opportunities associated with road infrastructure projects, enabling decision-makers to make more informed and resilient choices.

It is important to note that all the models developed by SYSTRA are prepared in accordance to the guidelines of the institutions responsible for validating the studies, guaranteeing their quality and reliability.

Road transport demand projection models play a fundamental role in planning, management and decision-making related to road infrastructure. In the context of highway concessions in Brazil, these models have a direct impact on public policies and private investment, influencing the economic viability and sustainability of projects.

SYSTRA stands out for its innovative approach to preparing demand projection models, considering a wide range of variables, and offering clients diverse scenarios. Combining advanced statistical analysis, consultation of international literature and adherence to regulatory guidelines, SYSTRA has contributed significantly to the advancement of the road transport sector in Brazil, promoting sustainable development and efficiency in infrastructure management.

Image credits: Getty Images

Don't miss anything, follow us on social media !