Early Warning System in the Prevention of Traffic Accidents

[lang_TR]

Dynamic Data Management Proposal for Early Warning System in the Prevention of Traffic Accidents: Case on İzmir’s Critical Streets

Achieving the goal of zero-accident on roads is at the agenda of EU. This urges more effective methods to help prevent further risk of accidents. First of all, it is required to estimate where, when and how the risk arises. Paralleling to the latest developments in the computer sciences and statistics, the studies on the accident risk estimation have increased in last ten years. Yet, the results of estimation models are not applicable to everywhere but rather site specific models are required due to the unknown parameters for obtaining healthy results. Traffic accident results have tenderly pointed to certain spots on roads, and this supports our hypothesis in the sense that black spots remain constant in time. The system proposed by the Project is rather a space-based schemata for data management process than a mathematical estimation model. Due to the wide use of intelligent of technologies in transportation (ITS), today, real time data can be manipulated so to provide a continuous information by which only data automation systems can handle. These do not require statistical model constructs. The ultimate purpose of the project is to maintain the police or traffic managers a system that could be used in the early warning sytem to alleviate accident risks. The system will be effective in readying for short-term operational measures, and to channellize the squads’ energy to right place and in right time when the risk arises. Besides, at will, even drivers can access the risk information on roads through which they drive. The aim of the project is to show how such a combination-based automation system can be accomplished by the method of category analysis, and to prove its validity via a pilot application of the system. Whether the approach produce precise results will be measured by the second year test results. Then the evalautions will be concluded for an video-image-based early warning system. The study will comprise of three basic stages, which will be completed in 30 months in total: Collecting the accident data and coding, checking the precision of categoric data system, evaluating the usefulness of the system for an early warning system.

TÜBİTAK-SOBAG 108K271


[/lang_TR][lang_en]

Dynamic Data Management Proposal for Early Warning System in the Prevention of Traffic Accidents: Case on İzmir’s Critical Streets

Achieving the goal of zero-accident on roads is at the agenda of EU. This urges more effective methods to help prevent further risk of accidents. First of all, it is required to estimate where, when and how the risk arises. Paralleling to the latest developments in the computer sciences and statistics, the studies on the accident risk estimation have increased in last ten years. Yet, the results of estimation models are not applicable to everywhere but rather site specific models are required due to the unknown parameters for obtaining healthy results. Traffic accident results have tenderly pointed to certain spots on roads, and this supports our hypothesis in the sense that black spots remain constant in time. The system proposed by the Project is rather a space-based schemata for data management process than a mathematical estimation model. Due to the wide use of intelligent of technologies in transportation (ITS), today, real time data can be manipulated so to provide a continuous information by which only data automation systems can handle. These do not require statistical model constructs. The ultimate purpose of the project is to maintain the police or traffic managers a system that could be used in the early warning sytem to alleviate accident risks. The system will be effective in readying for short-term operational measures, and to channellize the squads’ energy to right place and in right time when the risk arises. Besides, at will, even drivers can access the risk information on roads through which they drive. The aim of the project is to show how such a combination-based automation system can be accomplished by the method of category analysis, and to prove its validity via a pilot application of the system. Whether the approach produce precise results will be measured by the second year test results. Then the evalautions will be concluded for an video-image-based early warning system. The study will comprise of three basic stages, which will be completed in 30 months in total: Collecting the accident data and coding, checking the precision of categoric data system, evaluating the usefulness of the system for an early warning system.

TÜBİTAK-SOBAG 108K271

[/lang_en]

No comments