The development of a policy for road tax in Turkey, using a genetic algorithm approach for demand estimation Soner Haldenbilen, Halim Ceylan * Department of Civil Engineering, Engineering Faculty, Pamukkale University, Denizli 20017, Turkey Received 28 August 2004; received in revised form 23 February 2005; accepted 24 February 2005 Abstract Thisstudydeals,first,withestimationoftransportdemandbasedonGeneticAlgorithm(GA)approach, and then deals with the evaluation of the road tax system in Turkey. It proposes an alternative road tax policy. The total transport demand is estimated based on population, Gross Domestic Product per Capita (GDPPC), and vehicle-number. Three forms of the Genetic Algorithm Transport Demand Estimation for TaxRevenues(GATDETR)aredeveloped,ofwhichoneislinear,andthesecondandthirdareexponential forms of the mathematical expressions. The best-fit GATDETR model in terms of total minimum relative averageerrorsbetweenobservedandestimatedvaluesareselectedforfuturedemandestimation.Theeval- uation of the road tax system and policy proposal is made based on estimated demand. The Distance- Based-Taxation (DBT) system is proposed in order to control highway transport. With the DBT system, some road users may wish to use railway. Thus, we re-organize the railways in order to meet the demand, but this requires new fund. The DBT system may help to create to this fund. It may also help to develop fair-taxation for the road users. Results show that the GA can be used to model transport demand and henceincometaxinfuturetransportsplanning.Thisstudyalsosuggeststhatplanningthetaxationinhigh- way transport may help to ease funding problem of railway system. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Road tax; Transport demand; Genetic algorithm; Railway; Transport policy; Turkey 0965-8564/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.tra.2005.02.011 * Corresponding author. Tel.: +90 258 213 4030; fax: +90 258 212 5548. E-mail address: halimc@pamukkale.edu.tr (H. Ceylan). www.elsevier.com/locate/tra Transportation Research Part A 39 (2005) 861–877