• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.6
• /
• pp.3851-3857
• /
• 2014

Shippers' requests of freight trains vary with time, but generating an additional schedule of freight trains is not easy due to many considerations, such as the line capacity, operation rules, and conflicts with existing trains. On the other hand, an additional freight train schedule has been continuously requested and manually processed by domestic train operation companies using empirical method, which is time consuming. This paper proposes a model to determine the additional freight train schedule that assesses the feasibility of the added freight trains, and generates as many additional schedules as possible, while minimizing the delay of the existing schedules. The problem is presented using time-space network, modeled as multi-commodity flow problem, and solved using the column generation method. Three levels of experiment were conducted to show validity of the proposed model in the computation time.

• Proceedings of the KSR Conference
• /
• 2004.06a
• /
• pp.1374-1378
• /
• 2004

Railway conflict detection and resolution problem (RCDRP) involves complicated field constraints that should be considered for practical service. In this paper, we address those constraints in brief. Particularly, following situations are addressed; (1) temporal change of network topology, (2) consideration of diverse conditions of track and train, for example, single/double tracks and passenger/freight service, (3) siding capacity limitation, (4) bidirectional sides used by both inbound and outbound trains, (5) regulation for passenger transfer service, (6) consideration of siding length, (7) Restriction on stopping before the track segment with steep slope.

• Proceedings of the KIEE Conference
• /
• 2002.07d
• /
• pp.2736-2739
• /
• 2002

This paper is a study on the railway traffic conflict resolution(RTCR) using genetic algorithm(GA). Using GA, many complicated problems can be expressed in the easy way, and good solutions can be found in a reasonably short time. Due to the above merits, we focus to GA first of all. We express the sample RTCR problem and algorithm of Korean railroad using GA. From the Mark Ho's[2,3] results, we introduce a chromosome scheme in addition.

• Proceedings of the KSR Conference
• /
• 2007.05a
• /
• pp.1466-1474
• /
• 2007

In large station with high density traffic, trains is hardly controlled by CTC but by station dispatcher because CTC has difficulties in monitoring all states of affairs happening within the station such as departs and arrivals of many trains from different lines and shunting trains between yards and platforms, etc. Therefore station dispatcher has to make quick decision about how to schedule all trains including trains to arrive and depart before long. And such decisions become harder when an unexpected delay occurs. This study suggests a mathematical model and algorithm about how to detect in advance and resolve efficiently the conflicts occurring within a station and it will be shown how to reduce delay using our approach by means of analysing the schedule of ChyungRyangRi station.