• Journal of Korean Society of Transportation
• /
• v.20 no.4
• /
• pp.135-150
• /
• 2002

The basic assumption of analytical Dynamic Traffic Assignment models is that traffic demand and network conditions are known as a priori and unchanging during the whole planning horizon. This assumption may not be realistic in the practical traffic situation because traffic demand and network conditions nay vary from time to time. The rolling horizon implementation recognizes a fact : The Prediction of origin-destination(OD) matrices and network conditions is usually more accurate in a short period of time, while further into the whole horizon there exists a substantial uncertainty. In the rolling horizon implementation, therefore, rather than assuming time-dependent OD matrices and network conditions are known at the beginning of the horizon, it is assumed that the deterministic information of OD and traffic conditions for a short period are possessed, whereas information beyond this short period will not be available until the time rolls forward. This paper introduces rolling horizon implementation to enable a multi-class analytical DTA model to respond operationally to dynamic variations of both traffic demand and network conditions. In the paper, implementation procedure is discussed in detail, and practical solutions for some raised issues of 1) unfinished trips and 2) rerouting strategy of these trips, are proposed. Computational examples and results are presented and analyzed.

• Journal of Navigation and Port Research
• /
• v.38 no.3
• /
• pp.227-232
• /
• 2014

There are three essential components of eLoran: dLoran, data map of ASF, and the Loran data channel. Particularly, dLoran improves navigation accuracy, which is the core technology of eLoran systems. The requirement of HEA's absolute accuracy, less than 20 meters, can be satisfied via dLoran measurements and their corrections. In this study, dLoran measurements using the Pohang Loran-C (9930M) station signal were conducted at Yeongil Bay. We established a dLoran reference station at Homigot Management Office for navigation aids within the Bay. We estimated the effectiveness of the dLoran between the reference site (Homigot Management Office) and a test site (Heunghwan beach) by measuring TOAs. We verified that the TOA data measured at these two regions were highly correlated. The temporal differences in the data between the dLoran reference station and test site were about 10~30 ns per day, which is equivalent to a ranging error of 3~9 m. This result shows that eLoran can meet the requirement of 8~20 meters position accuracy for maritime HEA by correcting the ASF at the user's receiver.