Identification of risk areas for road traffic accidents using K-means clustering: a case study of Northern Thailand

Name: Kanwara Suwanthip

Organization : Thammasat University. Faculty of Science and Technology

Name: Supitchanut Chotpichet

Organization : Thammasat University. Faculty of Science and Technology

Email : amonrat.cho@dome.tu.ac.th

Name: Sangdao Wongsai

Organization : Thammasat University. Thammasat University Research Unit in Data Learning

Name: Noppachai Wongsai

Organization : Thammasat University. Thammasat University Research Unit in Data Learning

Name: Supet Jirakajohnkool

Organization : Thammasat University. Faculty of Science and Technology

keyword: Accident severity analysis

ThaSH: Traffic safety

; Point pattern analysis

ThaSH: Machine learning

ThaSH: Sustainable development

Abstract: In this study, we examined the risk areas of road traffic accidents (RTAs) in northern region of Thailand based on 3,870 points of RTAs in 2015. Forty-three clusters were identified using k-means clustering, a method of cluster analysis that partitions observed RTAs into K clusters in which each observation belongs to the cluster with the nearest centroid. The results showed two distribution patterns of accident points: a road network and a main road. RTAs within a road-network distribution were scattering around a cluster centroid within or between districts, depending on a cluster dispersion. This pattern was observed in Phrae and Uttaradit. A distribution pattern of RTAs along a straight-line road was found in Nan, Mae Hong Son, and Phayao. A greater dispersion signified a longer distance of RTAs from the centroid. At a region level, three largest densities of RTAs were all found in Chiangmai. The highest number of RTAs was 1,169, accounting for a density of 0.302, following by 292 (0.075), and 242 (0.063), respectively. A risk area was observed up north on the popular tourist road over three districts of Chiang Mai and further connecting to Chiang Rai. Other two risk areas were packed at the city centre and its adjacent districts. At a provincial level, Chiang Mai, Lamphun, and Nan had a density of greater than 0.60. Accident points in Chiang Mai and Lamphun were dense around their cluster centroid whereas those in Nan extended along the main road. Clustering risk areas of RTAs is beneficial for safety strategy before and after accidents. Relevant authorities could adopt our findings to identify associated factors for preventing RTAs at a particular road location and to establish a surveillance point to reach accident victims as quickly as possible for minimizing fatalities and injuries after crashes.

King Mongkut's University of Technology North Bangkok. Central Library

Address: BANGKOK

Email: library@kmutnb.ac.th

Created: 2022

Modified: 2025-09-16

Issued: 2025-09-16

บทความ/Article

application/pdf

BibliograpyCitation : In King Mongkut's University of Technology North Bangkok Faculty of Applied Science, Thai Statistical Association (TSA) and Statistics Cooperative Research Network (Statistics CRN). The Proceeding of International Conference on Applied Statistics (ICAS 2022) (pp.186-192). Bangkok : King Mongkut's University of Technology North Bangkok

eng

©copyrights King Mongkut's University of Technology North Bangkok

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	ICAS 2022pp.186-192.pdf	1.33 MB

ใช้เวลา

0.039182 วินาที