CONTACT US: Contact info
|
Citation |
Gal-Tzur A, Mahalel D, Prashker JN. Transp. Res. Rec. 1993; 1421: 69-75. |
|
Copyright |
(Copyright © 1993, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing) |
|
DOI |
unavailable |
PMID |
unavailable |
|
Abstract |
In many metropolitan areas traffic control is monitored from control centers. The operator of a control center is asked to make a quick decision and to modify the signal programs of the urban signal network in whole or in part. In order to make a proper decision, the operator must consider a wide range of alternatives and evaluate their expected effects on the whole traffic system. The complicated structure of the problem and the routine occurrence of random events demonstrate the complexity of the decision process in traffic control. A procedure will be described that is aimed at supporting this decision making. The procedure is characterized by the systematic scanning of a wide range of alternatives and includes a special algorithm for reducing the size of the problem and concentrating on the most promising strategies. A statistical decision tree is used for spanning all alternatives and expressing the subjective priorities among them and the projection regarding their consequences. An important option given to the controller is the ability to acquire more information to support his decision by using on-line simulations. This option is time consuming and therefore has a cost. The operator is given the tools to decide whether the additional information is worth the price. In addition, the system contains a systematic procedure to "learn" from past experience and to improve its ability to make decisions under uncertainty conditions. Language: en |
|
Keywords |
Costs; Mathematical models; Monitoring; Traffic signals; Decision support systems; Algorithms; Highway traffic control; Computer simulation; Decision theory; Trees (mathematics); Computational complexity |