Estimation of Cargo Demand at Major Seaports in India

Abstract

Estimation of cargo demand is essential for augmenting port capacity, planning port facilities, making decision on improvement of port operational efficiency, etc., and these activities need large and irreversible investment. Future cargo flow projections also helps in making decisions on cargo rates. Therefore, accuracy in estimating the cargo demand at port locations is critical. Indian ports experienced 151% increase in cargo volume between fiscal years from 2002-03 to 2014-15. This paper analyses such growth systematically with classical regression and time series modeling techniques. The regression models are developed using macroeconomic conditions as causal variables. The models are estimated using quarterly and monthly cargo flow data for twelve major seaports in India. The analysis revealed that the time series models perform better in terms of prediction accuracy than the regression models. The average prediction error from the regression models varied from 6% to 20%, while the error associations with time series models are varying from 3. 8% to 12.6%. This study is intended to provide infrastructure planners with some guidance on short to medium term development of transport infrastructure requirements over the lifetime of port infrastructure, while planning for port connectivity roads within an urban transport network.