The objective of this study is to estimate traffic growth using transport demand elasticity method and to compare how different these values are from the vehicle registration data.

In this present study, an attempt has been made to analyze the O-D data for passenger vehicles (cars & buses) and goods vehicle (trucks) collected by roadside interview method. The passenger characteristics such as average trip length, frequency and freight characteristics are analyzed and tabulated. The influence factors of various zones were found out. Socio-economic data viz -Per capita income and Net State Domestic Product, demographic data such as Population and registration of vehicle data of different states influencing the study stretch were collected from statistical data sources. The relationship between annual growths of vehicles in percentage over number of years is established.

To determine elasticity values, the regression analysis is carried out between socio economic variables growth index and vehicle growth index. The elasticity values for the future years are calculated based on the growth trend of vehicles.

The Project stretch is a part of NH-63 in the state of Karnataka which runs from east to west connecting Karnataka to Andhra Pradesh. The total length of NH-63 is about 432 km, out of which 370 km runs in Karnataka State and about 55.4 km runs in Andhra Pradesh.

This case study deals with Hubli – Hospet stretch of NH-63. The project stretch, starts at km 132+000 of NH-63 at junction with NH-4 Hubli-Dharwad bypass and ends at km 268+700 at junction of NH-63 and NH-13, Hitnal Junction.

For the purpose of forecasting traffic on the study stretch, several primary and secondary data were collected as mentioned below.

On the basis of reconnaissance survey and as per the recommendations of IRC, the project stretch was divided into two homogenous sections and suitable location for each section was strategically selected. The various traffic surveys carried out were:

• 7-day Continuous Traffic Volume Count [Table-1]
• 24 hour Origin and Destination Studies

Strict adherence to IRC codes and manuals were followed for the traffic surveys carried out.

• Fuel sales data along the study stretch
• Past data on traffic volume on the study stretch
• Previous year's vehicle registration data of Karnataka State
• Previous year's data on Per capita Income, Net State Domestic Product (NSDP), Population data of all the states influencing the project corridor and the National Average.

Table 1. Showing Average Daily Traffic

Traffic levels along a study stretch vary during different periods of time i.e., in different months/seasons. Information on this aspect is necessary to estimate the AADT (Table 3). This is best understood by studying monthly historical traffic volumes on the project corridor. This however is not available for the study stretch. In the absence of this direct information, it is customary to consider the monthly sales of petrol and diesel, at the fuel stations along the project corridor or on the road stretches in its environment. This information is presented in Table 2. The factors for passenger vehicles are based on petrol sales and that of goods vehicles (Trucks/LCV's) and buses on diesel sales.

Table 2. Showing daily fuel sales and Seasonal Correction Factors (SCF) required estimating AADT

Table 3. Showing AADT obtained after applying Seasonal Correction Factors

To establish the future traffic growth rates, following approaches have been explored.

• Past trends in Traffic growth on the Project Road.
• Growth of registered motor vehicles.
• Transport demand elasticity approach.

Past traffic data as collected from PWD is available for two locations (near Annigere and Gadag) along the project corridor. These data are available from January to July months of last 10 years. The growth rates were worked out for various categories of vehicles and conclusions were drawn.

Non-Uniformity in past traffic data of PWD may be attributed to errors during collection and processing of data and policy measures of the Government and other influences etc. To illustrate this point during recent years some of the mining activities around the project corridor have been banned by the Government which has caused a substantial decrease in the amount of trucks and Lorries. As the past traffic data on the Project Road is not showing any definite trend, one should not be guided by past traffic data for deriving growth rates.

An alternative approach is to explore the registered motor vehicles growth in the influence area and assume a growth rate equal to the average growth of vehicle registration. Such an assumption may not be correct, unless the area of influence is well defined and the general development pattern of influence area remains same. The growth rates for various modes are estimated and presented in Table 4. It can be observed from Table 4, during the last 7 years, average growth of two wheelers, cars and that of trucks is around 11%-12%. This high growth rate of more than 10% may not sustain in future. Therefore other rational approaches were explored in order to derive realistic growth rates.

Table 4. Summary of Cumulative Average Annual Growth Rate of Vehicles (%) in Karnataka state7

The exercise of traffic growth rate estimation has been carried out by us using the elasticity approach. The elasticity method relates traffic growth to changes in the related economic parameters. According to IRC-108-1996, elasticity based econometric model for highway projects could be derived in the following form:

Log e (P) = A0 + A1 Log e (EI)

Where:

• P = Traffic volume (of any vehicle type)
• EI= Economic Indicator (GDP/NSDP/Population/PCI)
• A0= Regression constant;
• A1= Regression co-efficient (Elasticity Index)

The main steps followed are:

• Defining the Project Influence Area from OD analysis of travel pattern.
• Estimating the past elasticity of traffic growth from time series of registered vehicles of influencing states.
• Assessment of future elasticity values for major vehicle groups, namely, cars, buses and trucks.

The results obtained from the Origin Destination surveys were used to identify the project influence area. The ratio of the total traffic originated/destined to a particular zone to the total traffic gives the influence factor for the particular zone. The influence factors were developed from the OD matrices and influence of each State is given in Table 5. A comparative study of the influence factors indicated that Karnataka State, where the project stretch runs has the majority influence of ninety two percent (92%). State of Goa, Andhra Pradesh and Maharashtra that has its border abutting Karnataka State has an influence factor of two percent (2%) four percent (4%) and two percent (2%) respectively. Tamil Nadu/Kerala and Rest of India has minimal or no share at all. These factors have all been accounted in derivation of the combined growth factor and utilized for the project sections.

Elasticity value is the factor by which the socio-economic growth rate is multiplied to get the growth rate of traffic. Traffic is directly linked to the economic growth such as per-capita income, population and NSDP/GDP. Considering the time series data on category wise registered vehicles and the economic variables, by regression analysis elasticity values is estimated as shown in Table 6.

Table 5. Influence of Vehicles observed on the Project Road

Table 6. Elasticity Values derived based on Regression Analysis for Karnataka State

Based on the moderated elasticity values and the projected economic/demographic indicators and with the given model as follows, the future average annual compound traffic growth rates by vehicle type are estimated.

Passenger Vehicles

Traffic Growth Rate = [ (1+rp) ( 1+ rpci x Em) – 1]

Where,

rp= Population Growth, rpci= Per capita Income Growth, Em= Elasticity

Goods Vehicles

Growth Rate for Goods Vehicles = Elasticity Value * NSDP Growth Rate

The growth rate estimated from elasticity values are shown in the Table 8 below:

Table 7. Showing adopted Elasticity values for future years

Table 8. Showing Growth Rates adopted for different classes of Vehicles

The comparison of growth rates on vehicular registration data and by elasticity value are as shown in Table below:

• The Growth estimates based on past traffic data collected from PWD department was found to be inadequate and there was no definite growth trend observed. Table 9 provides details of the Growth rates of various Methods adopted for the study.
• It can be observed from Table 9, during the last 7 years, average growth of two wheelers, cars and that of trucks is around 11%-12%. This high growth rate of more than 10% may not sustain in future. Therefore other rational approach i.e. elasticity method was adopted in order to derive realistic growth rates.
• The Growth rates obtained by elasticity method were adopted for traffic forecasting and the Optimistic growth rates were adopted as the project corridor in and around places where there is a huge Economic development taking place.
• In the area of influence of the project road, there is no good network of National Highways. The development of the project corridor will further boost the economy of the area and is considered to be the most desirable feature for the social and economical prosperity for the people of the area.
• For the purpose of highway design and project development, we need to adopt one Scenario. The present, optimistic, economic outlook is based on the experience of only past few years. Also, with structural changes in the economy the traffic elasticities with respect to economic growth may go down more than expected.
• High fuel costs may also affect traffic growth negatively. However, it is understood that Karnataka as a state has not yet reached its socio-economic potential and would therefore register a higher growth impetus. Further, the project corridor connects to one of the largest iron ore reserve in India and some important cities like Karwar and Hubli.
• Although, it is difficult to translate macro-economic indicators at a micro-economic regional level, we believe the growth of this region would be faster than the general growth rate of the state. In view of the above discussion Optimistic Growth Rates has been adopted for Traffic Forecasting.
• It is observed that with the constraints of availability of proper data and fluctuating developing economy, the task of Traffic Demand Forecasting could be quite tricky, subjective and approximate.

The growth rates obtained from transport demand elasticity method is a widely used method all over India. These growth rates are adopted to predict future traffic volumes and Laning requirements.

Table 9. Comparison of Growth Rates by Various Methods