UPORABA PROSTORSKE STATISTIKE ZA DOLOČEVANJE ZGOSTITEV PROMETNIH NESREČ
IDENTIFICATION OF ROAD ACCIDENT HOT SPOTS USING SPATIAL STATISTICS

Peter Lipar, Jure Kostanjšek, Marijan Žura

DOI: 10.15292/geodetski-vestnik.2010.01.061-069

 

Izvleček:

V prispevku predstavljamo prilagojeno metodo za odkrivanje nevarnih odsekov in krajev, ki temelji na uporabi indeksa Getis-Ord G*. Od klasične metode se razlikuje v tem, da poleg teže nezgod upoštevamo prometne obremenitve na kraju nezgode. Metoda je bila uporabljena na primeru omrežja slovenskih avtocest in hitrih cest. Prikazani so rezultati analiz z upoštevanjem prometnih obremenitev in brez njih oziroma primerjava z metodama določanja gostote jeder in lokalnega Anselin-Moranovega indeksa.

Ključne besede: cestnoprometne nezgode, črne točke, prostorska statistika

 

Abstract:

This paper presents a modified spatial statistical method for identifying road accident hot spots. Our suggested method is based on the Getis-Ord G* index. It differs from the standard method, because in addition to the number of accidents and their consequences, we also consider traffic at the locations of these accidents. The method was tested on the Slovene motorway network. The results are presented with and without considering traffic volumes and are compared with both Kernel density method and Anselin-Moran local index method.

Keywords: road traffic accidents, black spots, spatial statistics

 

Literatura / References:

Anderson, T. K. (2009). Kernel density estimation and K-means clustering to profile road accident hotspots. Accident Analysis and Prevention, 41(3), 359–364.
http://dx.doi.org/10.1016/j.aap.2008.12.014

Cheng, W., Washington, S. (2008). New Criteria for Evaluating Methods of Identifying Hot Spots. Journal of the Transportation Research Board, (2083), 76–85.
http://dx.doi.org/10.3141/2083-09

Direkcija RS za ceste, Omega Consult d. o. o. (2009). Določitev nevarnih mest na državnem cestnem omrežju v Republiki Sloveniji. Končno poročilo.

Moons, E., Brijs, T., Wets, G. (2008). Hot Spot Analysis: Improving a Local Indicator of Spatial Association for KE Application in Traffic Safety. Lecture Notes in Computer Science: Computational Science and Its Applications – ICCSA.

Moons, E., Brijs, T., Wets, G. (2009). Improving Moran’s Index to Identify Hot Spots in Traffic Safety. Studies in Computational Intelligence Geocomputation and Urban Planning.
http://dx.doi.org/10.1007/978-3-540-89930-3_7

Páez, A., Scott, D. M. (2005). Spatial statistics for urban analysis: A review of techniques with examples. GeoJournal, 61(1), 53–67.
http://dx.doi.org/10.1007/s10708-005-0877-5

Sugihara, K., Okabe, A., Satoh, T. (2009). Computational method for the point cluster analysis on networks. Geoinformatica Online.
http://dx.doi.org/10.1007/s10707-009-0092-5

Xie, Z., Yan, J. (2008). Kernel density estimation of traffic accidents in a network space. Computers, Environment and Urban Systems, 32, 396–406.
http://dx.doi.org/10.1016/j.compenvurbsys.2008.05.001