NEURAL NETWORK BASED MODEL FOR PREDICTING THE NUMBER OF SLEEPING CARS IN RAIL TRANSPORT

Journal Title: INTERNATIONAL JOURNAL FOR TRAFFIC AND TRANSPORT ENGINEERING - Year 2015, Vol 5, Issue 1

Abstract

A Decision Support System based on Artificial Neural Network is developed to forecast the number of sleeping cars in rail transport. The inputs to the system consist of train route, month, type of sleeping car, number of berths (supply), number of departures, ticket price and GDP, while the output of the neural network is the number of sold tickets (demand). By comparing the results obtained through the model with those resulting from historical data, it has been found that the developed model is highly compatible with reality. The developed Decision Support System could be used for capacity planning purposes, because it is important for a rail operator to know in advance how many sleeping cars have to be available. All considered data are obtained from Serbian Railways.

Authors and Affiliations

Dragana Macura, Milica Šelmić, Branka Dimitrijević, Milorad Miletić

Keywords

Related Articles

A MODEL TO ESTIMATE THE PASSENGER RAIL LIBERALISATION: THE CASE OF SERBIA

Passenger rail transport market share has been drastically reduced in the last fifty years. The reforms were aimed at opening the market of railway services and development of competition in the railway sector in order t...

FORECASTING AND INVENTORY PERFORMANCE IN DIRECT-STORE DELIVERY SUPPLY CHAIN: CASE OF RETAILER IN SERBIA

This paper presents research conducted on the case of food retailer in Serbia in order to answer at which level are the forecasting techniques applicable to such companies, at given supply chain model and in given enviro...

RISK FACTORS AFFECTING FATAL VERSUS NON-FATAL ROAD TRAFFIC ACCIDENTS: THE CASE OF KARS PROVINCE, TURKEY

The aim of this paper is to determine risk factors affecting the fatal versus non-fatal accidents in a rural region of Turkey, during 2008-2012, considering variables associated with the individual, the environment, and...

RAZVOJ I PROCENA PROTOTIPNOG SISTEMA SVETLOSNIH OZNAKA NA OGRADAMA

Trenutno se na ogradama u zonama radova koriste standardna svetla upozorenja koja imaju funkcije usmeravanja i upozoravanja. Ova trepćuća žuta svetla u zonama radova danas se koriste u različite svrhe. Razvijeni su i pro...

COST - BENEFIT ASSESSMENT OF AIRCRAFT SAFETY

The constant increase of air traffic volume causes the requirements for development and implementation of numerous measures that will enable monitoring of the safety level in the aviation organizations. An inseparable el...

Download PDF file
  • EP ID EP132296
  • DOI 10.7708/ijtte.2015.5(1).04
  • Views 74
  • Downloads 0

How To Cite

Dragana Macura, Milica Šelmić, Branka Dimitrijević, Milorad Miletić (2015). NEURAL NETWORK BASED MODEL FOR PREDICTING THE NUMBER OF SLEEPING CARS IN RAIL TRANSPORT. INTERNATIONAL JOURNAL FOR TRAFFIC AND TRANSPORT ENGINEERING, 5(1), 29-35. https://europub.co.uk/articles/-A-132296